diff --git "a/data/benchmarks/bfcl.json" "b/data/benchmarks/bfcl.json"
deleted file mode 100644--- "a/data/benchmarks/bfcl.json"
+++ /dev/null
@@ -1,4117 +0,0 @@
-{
-  "models": [
-    {
-      "model_id": "amazon/amazon-nova-2-lite-v1-0-fc",
-      "name": "Amazon-Nova-2-Lite-v1:0 (FC)",
-      "developer": "amazon",
-      "scores": {
-        "bfcl.overall.rank": 80.0,
-        "bfcl.overall.overall_accuracy": 27.1,
-        "bfcl.overall.total_cost_usd": 78.19,
-        "bfcl.overall.latency_mean_s": 8.55,
-        "bfcl.overall.latency_std_s": 9.85,
-        "bfcl.overall.latency_p95_s": 27.62,
-        "bfcl.non_live.ast_accuracy": 86.96,
-        "bfcl.non_live.simple_ast_accuracy": 76.33,
-        "bfcl.non_live.multiple_ast_accuracy": 94.0,
-        "bfcl.non_live.parallel_ast_accuracy": 91.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 86.0,
-        "bfcl.live.live_accuracy": 80.83,
-        "bfcl.live.live_simple_ast_accuracy": 83.33,
-        "bfcl.live.live_multiple_ast_accuracy": 80.15,
-        "bfcl.live.live_parallel_ast_accuracy": 87.5,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 79.17,
-        "bfcl.multi_turn.accuracy": 2.12,
-        "bfcl.multi_turn.base_accuracy": 2.5,
-        "bfcl.multi_turn.miss_function_accuracy": 1.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 2.0,
-        "bfcl.multi_turn.long_context_accuracy": 2.5,
-        "bfcl.web_search.accuracy": 5.0,
-        "bfcl.web_search.base_accuracy": 4.0,
-        "bfcl.web_search.no_snippet_accuracy": 6.0,
-        "bfcl.memory.accuracy": 2.37,
-        "bfcl.memory.kv_accuracy": 1.94,
-        "bfcl.memory.vector_accuracy": 3.23,
-        "bfcl.memory.recursive_summarization_accuracy": 1.94,
-        "bfcl.relevance.relevance_detection_accuracy": 75.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 82.11
-      }
-    },
-    {
-      "model_id": "amazon/amazon-nova-micro-v1-0-fc",
-      "name": "Amazon-Nova-Micro-v1:0 (FC)",
-      "developer": "amazon",
-      "scores": {
-        "bfcl.overall.rank": 95.0,
-        "bfcl.overall.overall_accuracy": 22.29,
-        "bfcl.overall.total_cost_usd": 1.81,
-        "bfcl.overall.latency_mean_s": 1.12,
-        "bfcl.overall.latency_std_s": 0.45,
-        "bfcl.overall.latency_p95_s": 1.79,
-        "bfcl.non_live.ast_accuracy": 74.1,
-        "bfcl.non_live.simple_ast_accuracy": 70.92,
-        "bfcl.non_live.multiple_ast_accuracy": 87.5,
-        "bfcl.non_live.parallel_ast_accuracy": 75.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 62.5,
-        "bfcl.live.live_accuracy": 66.32,
-        "bfcl.live.live_simple_ast_accuracy": 72.09,
-        "bfcl.live.live_multiple_ast_accuracy": 64.96,
-        "bfcl.live.live_parallel_ast_accuracy": 81.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 54.17,
-        "bfcl.multi_turn.accuracy": 1.38,
-        "bfcl.multi_turn.base_accuracy": 1.5,
-        "bfcl.multi_turn.miss_function_accuracy": 1.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 2.0,
-        "bfcl.multi_turn.long_context_accuracy": 1.0,
-        "bfcl.web_search.accuracy": 1.5,
-        "bfcl.web_search.base_accuracy": 3.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 2.37,
-        "bfcl.memory.kv_accuracy": 1.94,
-        "bfcl.memory.vector_accuracy": 2.58,
-        "bfcl.memory.recursive_summarization_accuracy": 2.58,
-        "bfcl.relevance.relevance_detection_accuracy": 81.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 70.65
-      }
-    },
-    {
-      "model_id": "amazon/amazon-nova-pro-v1-0-fc",
-      "name": "Amazon-Nova-Pro-v1:0 (FC)",
-      "developer": "amazon",
-      "scores": {
-        "bfcl.overall.rank": 88.0,
-        "bfcl.overall.overall_accuracy": 24.97,
-        "bfcl.overall.total_cost_usd": 48.44,
-        "bfcl.overall.latency_mean_s": 2.25,
-        "bfcl.overall.latency_std_s": 1.91,
-        "bfcl.overall.latency_p95_s": 3.29,
-        "bfcl.non_live.ast_accuracy": 86.58,
-        "bfcl.non_live.simple_ast_accuracy": 75.83,
-        "bfcl.non_live.multiple_ast_accuracy": 93.0,
-        "bfcl.non_live.parallel_ast_accuracy": 93.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 84.0,
-        "bfcl.live.live_accuracy": 78.53,
-        "bfcl.live.live_simple_ast_accuracy": 81.4,
-        "bfcl.live.live_multiple_ast_accuracy": 77.97,
-        "bfcl.live.live_parallel_ast_accuracy": 81.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 70.83,
-        "bfcl.multi_turn.accuracy": 1.88,
-        "bfcl.multi_turn.base_accuracy": 1.5,
-        "bfcl.multi_turn.miss_function_accuracy": 0.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 2.5,
-        "bfcl.multi_turn.long_context_accuracy": 3.0,
-        "bfcl.web_search.accuracy": 2.5,
-        "bfcl.web_search.base_accuracy": 4.0,
-        "bfcl.web_search.no_snippet_accuracy": 1.0,
-        "bfcl.memory.accuracy": 1.94,
-        "bfcl.memory.kv_accuracy": 2.58,
-        "bfcl.memory.vector_accuracy": 1.29,
-        "bfcl.memory.recursive_summarization_accuracy": 1.94,
-        "bfcl.relevance.relevance_detection_accuracy": 93.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 70.06
-      }
-    },
-    {
-      "model_id": "anthropic/claude-haiku-4-5-20251001-fc",
-      "name": "Claude-Haiku-4-5-20251001 (FC)",
-      "developer": "Anthropic",
-      "scores": {
-        "bfcl.overall.rank": 6.0,
-        "bfcl.overall.overall_accuracy": 68.7,
-        "bfcl.overall.total_cost_usd": 14.23,
-        "bfcl.overall.latency_mean_s": 1.68,
-        "bfcl.overall.latency_std_s": 3.92,
-        "bfcl.overall.latency_p95_s": 3.15,
-        "bfcl.non_live.ast_accuracy": 86.5,
-        "bfcl.non_live.simple_ast_accuracy": 71.0,
-        "bfcl.non_live.multiple_ast_accuracy": 94.0,
-        "bfcl.non_live.parallel_ast_accuracy": 92.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 88.5,
-        "bfcl.live.live_accuracy": 78.68,
-        "bfcl.live.live_simple_ast_accuracy": 83.72,
-        "bfcl.live.live_multiple_ast_accuracy": 77.59,
-        "bfcl.live.live_parallel_ast_accuracy": 75.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 75.0,
-        "bfcl.multi_turn.accuracy": 53.62,
-        "bfcl.multi_turn.base_accuracy": 63.5,
-        "bfcl.multi_turn.miss_function_accuracy": 42.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 52.5,
-        "bfcl.multi_turn.long_context_accuracy": 56.0,
-        "bfcl.web_search.accuracy": 83.5,
-        "bfcl.web_search.base_accuracy": 86.0,
-        "bfcl.web_search.no_snippet_accuracy": 81.0,
-        "bfcl.memory.accuracy": 54.41,
-        "bfcl.memory.kv_accuracy": 51.61,
-        "bfcl.memory.vector_accuracy": 55.48,
-        "bfcl.memory.recursive_summarization_accuracy": 56.13,
-        "bfcl.relevance.relevance_detection_accuracy": 62.5,
-        "bfcl.relevance.irrelevance_detection_accuracy": 85.11
-      }
-    },
-    {
-      "model_id": "anthropic/claude-haiku-4-5-20251001-prompt",
-      "name": "Claude-Haiku-4-5-20251001 (Prompt)",
-      "developer": "Anthropic",
-      "scores": {
-        "bfcl.overall.rank": 87.0,
-        "bfcl.overall.overall_accuracy": 25.26,
-        "bfcl.overall.total_cost_usd": 45.13,
-        "bfcl.overall.latency_mean_s": 3.75,
-        "bfcl.overall.latency_std_s": 19.96,
-        "bfcl.overall.latency_p95_s": 3.77,
-        "bfcl.non_live.ast_accuracy": 55.42,
-        "bfcl.non_live.simple_ast_accuracy": 55.67,
-        "bfcl.non_live.multiple_ast_accuracy": 84.0,
-        "bfcl.non_live.parallel_ast_accuracy": 38.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 44.0,
-        "bfcl.live.live_accuracy": 52.48,
-        "bfcl.live.live_simple_ast_accuracy": 66.67,
-        "bfcl.live.live_multiple_ast_accuracy": 49.76,
-        "bfcl.live.live_parallel_ast_accuracy": 56.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 16.67,
-        "bfcl.multi_turn.accuracy": 1.75,
-        "bfcl.multi_turn.base_accuracy": 1.5,
-        "bfcl.multi_turn.miss_function_accuracy": 0.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 4.0,
-        "bfcl.multi_turn.long_context_accuracy": 1.5,
-        "bfcl.web_search.accuracy": 19.5,
-        "bfcl.web_search.base_accuracy": 20.0,
-        "bfcl.web_search.no_snippet_accuracy": 19.0,
-        "bfcl.memory.accuracy": 2.58,
-        "bfcl.memory.kv_accuracy": 2.58,
-        "bfcl.memory.vector_accuracy": 1.94,
-        "bfcl.memory.recursive_summarization_accuracy": 3.23,
-        "bfcl.relevance.relevance_detection_accuracy": 31.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 95.29,
-        "bfcl.format_sensitivity.max_delta": 67.5,
-        "bfcl.format_sensitivity.stddev": 20.07
-      }
-    },
-    {
-      "model_id": "anthropic/claude-opus-4-5-20251101-fc",
-      "name": "Claude-Opus-4-5-20251101 (FC)",
-      "developer": "Anthropic",
-      "scores": {
-        "bfcl.overall.rank": 1.0,
-        "bfcl.overall.overall_accuracy": 77.47,
-        "bfcl.overall.total_cost_usd": 86.55,
-        "bfcl.overall.latency_mean_s": 4.38,
-        "bfcl.overall.latency_std_s": 3.13,
-        "bfcl.overall.latency_p95_s": 7.56,
-        "bfcl.non_live.ast_accuracy": 88.58,
-        "bfcl.non_live.simple_ast_accuracy": 76.83,
-        "bfcl.non_live.multiple_ast_accuracy": 95.5,
-        "bfcl.non_live.parallel_ast_accuracy": 93.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 88.5,
-        "bfcl.live.live_accuracy": 79.79,
-        "bfcl.live.live_simple_ast_accuracy": 86.43,
-        "bfcl.live.live_multiple_ast_accuracy": 78.16,
-        "bfcl.live.live_parallel_ast_accuracy": 87.5,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 75.0,
-        "bfcl.multi_turn.accuracy": 68.38,
-        "bfcl.multi_turn.base_accuracy": 81.0,
-        "bfcl.multi_turn.miss_function_accuracy": 64.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 58.0,
-        "bfcl.multi_turn.long_context_accuracy": 70.5,
-        "bfcl.web_search.accuracy": 84.5,
-        "bfcl.web_search.base_accuracy": 84.0,
-        "bfcl.web_search.no_snippet_accuracy": 85.0,
-        "bfcl.memory.accuracy": 73.76,
-        "bfcl.memory.kv_accuracy": 70.97,
-        "bfcl.memory.vector_accuracy": 72.9,
-        "bfcl.memory.recursive_summarization_accuracy": 77.42,
-        "bfcl.relevance.relevance_detection_accuracy": 62.5,
-        "bfcl.relevance.irrelevance_detection_accuracy": 84.72
-      }
-    },
-    {
-      "model_id": "anthropic/claude-opus-4-5-20251101-prompt",
-      "name": "Claude-Opus-4-5-20251101 (Prompt)",
-      "developer": "Anthropic",
-      "scores": {
-        "bfcl.overall.rank": 57.0,
-        "bfcl.overall.overall_accuracy": 33.47,
-        "bfcl.overall.total_cost_usd": 88.33,
-        "bfcl.overall.latency_mean_s": 3.76,
-        "bfcl.overall.latency_std_s": 13.19,
-        "bfcl.overall.latency_p95_s": 5.52,
-        "bfcl.non_live.ast_accuracy": 89.65,
-        "bfcl.non_live.simple_ast_accuracy": 79.58,
-        "bfcl.non_live.multiple_ast_accuracy": 93.5,
-        "bfcl.non_live.parallel_ast_accuracy": 93.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 92.5,
-        "bfcl.live.live_accuracy": 76.02,
-        "bfcl.live.live_simple_ast_accuracy": 84.5,
-        "bfcl.live.live_multiple_ast_accuracy": 74.17,
-        "bfcl.live.live_parallel_ast_accuracy": 81.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 62.5,
-        "bfcl.multi_turn.accuracy": 16.12,
-        "bfcl.multi_turn.base_accuracy": 20.5,
-        "bfcl.multi_turn.miss_function_accuracy": 9.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 21.5,
-        "bfcl.multi_turn.long_context_accuracy": 13.5,
-        "bfcl.web_search.accuracy": 13.0,
-        "bfcl.web_search.base_accuracy": 13.0,
-        "bfcl.web_search.no_snippet_accuracy": 13.0,
-        "bfcl.memory.accuracy": 1.94,
-        "bfcl.memory.kv_accuracy": 1.29,
-        "bfcl.memory.vector_accuracy": 1.94,
-        "bfcl.memory.recursive_summarization_accuracy": 2.58,
-        "bfcl.relevance.relevance_detection_accuracy": 68.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 90.75,
-        "bfcl.format_sensitivity.max_delta": 13.0,
-        "bfcl.format_sensitivity.stddev": 3.65
-      }
-    },
-    {
-      "model_id": "anthropic/claude-sonnet-4-5-20250929-fc",
-      "name": "Claude-Sonnet-4-5-20250929 (FC)",
-      "developer": "Anthropic",
-      "scores": {
-        "bfcl.overall.rank": 2.0,
-        "bfcl.overall.overall_accuracy": 73.24,
-        "bfcl.overall.total_cost_usd": 43.73,
-        "bfcl.overall.latency_mean_s": 4.31,
-        "bfcl.overall.latency_std_s": 4.43,
-        "bfcl.overall.latency_p95_s": 7.27,
-        "bfcl.non_live.ast_accuracy": 88.65,
-        "bfcl.non_live.simple_ast_accuracy": 72.58,
-        "bfcl.non_live.multiple_ast_accuracy": 95.5,
-        "bfcl.non_live.parallel_ast_accuracy": 94.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 92.0,
-        "bfcl.live.live_accuracy": 81.13,
-        "bfcl.live.live_simple_ast_accuracy": 89.53,
-        "bfcl.live.live_multiple_ast_accuracy": 78.92,
-        "bfcl.live.live_parallel_ast_accuracy": 87.5,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 83.33,
-        "bfcl.multi_turn.accuracy": 61.37,
-        "bfcl.multi_turn.base_accuracy": 69.0,
-        "bfcl.multi_turn.miss_function_accuracy": 65.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 52.5,
-        "bfcl.multi_turn.long_context_accuracy": 59.0,
-        "bfcl.web_search.accuracy": 81.0,
-        "bfcl.web_search.base_accuracy": 82.0,
-        "bfcl.web_search.no_snippet_accuracy": 80.0,
-        "bfcl.memory.accuracy": 64.95,
-        "bfcl.memory.kv_accuracy": 54.19,
-        "bfcl.memory.vector_accuracy": 57.42,
-        "bfcl.memory.recursive_summarization_accuracy": 83.23,
-        "bfcl.relevance.relevance_detection_accuracy": 68.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 86.61
-      }
-    },
-    {
-      "model_id": "anthropic/claude-sonnet-4-5-20250929-prompt",
-      "name": "Claude-Sonnet-4-5-20250929 (Prompt)",
-      "developer": "Anthropic",
-      "scores": {
-        "bfcl.overall.rank": 89.0,
-        "bfcl.overall.overall_accuracy": 24.9,
-        "bfcl.overall.total_cost_usd": 47.82,
-        "bfcl.overall.latency_mean_s": 3.84,
-        "bfcl.overall.latency_std_s": 1.53,
-        "bfcl.overall.latency_p95_s": 6.66,
-        "bfcl.non_live.ast_accuracy": 59.81,
-        "bfcl.non_live.simple_ast_accuracy": 47.25,
-        "bfcl.non_live.multiple_ast_accuracy": 79.5,
-        "bfcl.non_live.parallel_ast_accuracy": 53.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 59.0,
-        "bfcl.live.live_accuracy": 46.56,
-        "bfcl.live.live_simple_ast_accuracy": 73.26,
-        "bfcl.live.live_multiple_ast_accuracy": 40.17,
-        "bfcl.live.live_parallel_ast_accuracy": 56.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 33.33,
-        "bfcl.multi_turn.accuracy": 1.62,
-        "bfcl.multi_turn.base_accuracy": 2.0,
-        "bfcl.multi_turn.miss_function_accuracy": 0.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 3.0,
-        "bfcl.multi_turn.long_context_accuracy": 1.5,
-        "bfcl.web_search.accuracy": 16.0,
-        "bfcl.web_search.base_accuracy": 16.0,
-        "bfcl.web_search.no_snippet_accuracy": 16.0,
-        "bfcl.memory.accuracy": 5.38,
-        "bfcl.memory.kv_accuracy": 4.52,
-        "bfcl.memory.vector_accuracy": 9.68,
-        "bfcl.memory.recursive_summarization_accuracy": 1.94,
-        "bfcl.relevance.relevance_detection_accuracy": 37.5,
-        "bfcl.relevance.irrelevance_detection_accuracy": 95.03,
-        "bfcl.format_sensitivity.max_delta": 37.5,
-        "bfcl.format_sensitivity.stddev": 10.07
-      }
-    },
-    {
-      "model_id": "bittensor/bitagent-bounty-8b",
-      "name": "BitAgent-Bounty-8B",
-      "developer": "bittensor",
-      "scores": {
-        "bfcl.overall.rank": 36.0,
-        "bfcl.overall.overall_accuracy": 46.23,
-        "bfcl.overall.total_cost_usd": 18.02,
-        "bfcl.overall.latency_mean_s": 16.52,
-        "bfcl.overall.latency_std_s": 30.73,
-        "bfcl.overall.latency_p95_s": 77.12,
-        "bfcl.non_live.ast_accuracy": 81.6,
-        "bfcl.non_live.simple_ast_accuracy": 72.42,
-        "bfcl.non_live.multiple_ast_accuracy": 93.0,
-        "bfcl.non_live.parallel_ast_accuracy": 83.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 78.0,
-        "bfcl.live.live_accuracy": 93.12,
-        "bfcl.live.live_simple_ast_accuracy": 90.31,
-        "bfcl.live.live_multiple_ast_accuracy": 94.02,
-        "bfcl.live.live_parallel_ast_accuracy": 75.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 95.83,
-        "bfcl.multi_turn.accuracy": 62.38,
-        "bfcl.multi_turn.base_accuracy": 75.0,
-        "bfcl.multi_turn.miss_function_accuracy": 49.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 68.0,
-        "bfcl.multi_turn.long_context_accuracy": 57.0,
-        "bfcl.web_search.accuracy": 0.0,
-        "bfcl.web_search.base_accuracy": 0.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 1.51,
-        "bfcl.memory.kv_accuracy": 1.29,
-        "bfcl.memory.vector_accuracy": 1.29,
-        "bfcl.memory.recursive_summarization_accuracy": 1.94,
-        "bfcl.relevance.relevance_detection_accuracy": 68.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 97.48
-      }
-    },
-    {
-      "model_id": "cohere/command-a-fc",
-      "name": "Command A (FC)",
-      "developer": "cohere",
-      "scores": {
-        "bfcl.overall.rank": 35.0,
-        "bfcl.overall.overall_accuracy": 46.49,
-        "bfcl.overall.total_cost_usd": 91.37,
-        "bfcl.overall.latency_mean_s": 2.09,
-        "bfcl.overall.latency_std_s": 7.36,
-        "bfcl.overall.latency_p95_s": 4.94,
-        "bfcl.non_live.ast_accuracy": 87.56,
-        "bfcl.non_live.simple_ast_accuracy": 75.75,
-        "bfcl.non_live.multiple_ast_accuracy": 93.0,
-        "bfcl.non_live.parallel_ast_accuracy": 93.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 88.0,
-        "bfcl.live.live_accuracy": 78.53,
-        "bfcl.live.live_simple_ast_accuracy": 85.66,
-        "bfcl.live.live_multiple_ast_accuracy": 76.92,
-        "bfcl.live.live_parallel_ast_accuracy": 81.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 70.83,
-        "bfcl.multi_turn.accuracy": 29.5,
-        "bfcl.multi_turn.base_accuracy": 38.0,
-        "bfcl.multi_turn.miss_function_accuracy": 23.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 32.0,
-        "bfcl.multi_turn.long_context_accuracy": 25.0,
-        "bfcl.web_search.accuracy": 46.5,
-        "bfcl.web_search.base_accuracy": 60.0,
-        "bfcl.web_search.no_snippet_accuracy": 33.0,
-        "bfcl.memory.accuracy": 16.56,
-        "bfcl.memory.kv_accuracy": 4.52,
-        "bfcl.memory.vector_accuracy": 5.16,
-        "bfcl.memory.recursive_summarization_accuracy": 40.0,
-        "bfcl.relevance.relevance_detection_accuracy": 81.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 84.19
-      }
-    },
-    {
-      "model_id": "cohere/command-a-reasoning-fc",
-      "name": "Command A Reasoning (FC)",
-      "developer": "cohere",
-      "scores": {
-        "bfcl.overall.rank": 13.0,
-        "bfcl.overall.overall_accuracy": 57.06,
-        "bfcl.overall.total_cost_usd": 3.04,
-        "bfcl.overall.latency_mean_s": 3.44,
-        "bfcl.overall.latency_std_s": 4.91,
-        "bfcl.overall.latency_p95_s": 8.39,
-        "bfcl.non_live.ast_accuracy": 86.27,
-        "bfcl.non_live.simple_ast_accuracy": 73.58,
-        "bfcl.non_live.multiple_ast_accuracy": 93.5,
-        "bfcl.non_live.parallel_ast_accuracy": 89.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 88.5,
-        "bfcl.live.live_accuracy": 78.61,
-        "bfcl.live.live_simple_ast_accuracy": 80.23,
-        "bfcl.live.live_multiple_ast_accuracy": 78.35,
-        "bfcl.live.live_parallel_ast_accuracy": 75.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 75.0,
-        "bfcl.multi_turn.accuracy": 50.12,
-        "bfcl.multi_turn.base_accuracy": 61.5,
-        "bfcl.multi_turn.miss_function_accuracy": 41.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 49.5,
-        "bfcl.multi_turn.long_context_accuracy": 48.5,
-        "bfcl.web_search.accuracy": 55.5,
-        "bfcl.web_search.base_accuracy": 65.0,
-        "bfcl.web_search.no_snippet_accuracy": 46.0,
-        "bfcl.memory.accuracy": 28.82,
-        "bfcl.memory.kv_accuracy": 16.13,
-        "bfcl.memory.vector_accuracy": 23.87,
-        "bfcl.memory.recursive_summarization_accuracy": 46.45,
-        "bfcl.relevance.relevance_detection_accuracy": 68.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 86.75
-      }
-    },
-    {
-      "model_id": "cohere/command-r7b-fc",
-      "name": "Command R7B (FC)",
-      "developer": "cohere",
-      "scores": {
-        "bfcl.overall.rank": 61.0,
-        "bfcl.overall.overall_accuracy": 32.07,
-        "bfcl.overall.total_cost_usd": 1.5,
-        "bfcl.overall.latency_mean_s": 1.38,
-        "bfcl.overall.latency_std_s": 2.87,
-        "bfcl.overall.latency_p95_s": 2.69,
-        "bfcl.non_live.ast_accuracy": 80.96,
-        "bfcl.non_live.simple_ast_accuracy": 67.33,
-        "bfcl.non_live.multiple_ast_accuracy": 89.5,
-        "bfcl.non_live.parallel_ast_accuracy": 85.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 81.5,
-        "bfcl.live.live_accuracy": 69.06,
-        "bfcl.live.live_simple_ast_accuracy": 62.79,
-        "bfcl.live.live_multiple_ast_accuracy": 70.94,
-        "bfcl.live.live_parallel_ast_accuracy": 43.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 70.83,
-        "bfcl.multi_turn.accuracy": 8.25,
-        "bfcl.multi_turn.base_accuracy": 12.0,
-        "bfcl.multi_turn.miss_function_accuracy": 0.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 10.5,
-        "bfcl.multi_turn.long_context_accuracy": 10.0,
-        "bfcl.web_search.accuracy": 27.0,
-        "bfcl.web_search.base_accuracy": 43.0,
-        "bfcl.web_search.no_snippet_accuracy": 11.0,
-        "bfcl.memory.accuracy": 5.16,
-        "bfcl.memory.kv_accuracy": 2.58,
-        "bfcl.memory.vector_accuracy": 9.68,
-        "bfcl.memory.recursive_summarization_accuracy": 3.23,
-        "bfcl.relevance.relevance_detection_accuracy": 68.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 81.65
-      }
-    },
-    {
-      "model_id": "deepseek/deepseek-v3-2-exp-fc",
-      "name": "DeepSeek-V3.2-Exp (FC)",
-      "developer": "deepseek",
-      "scores": {
-        "bfcl.overall.rank": 19.0,
-        "bfcl.overall.overall_accuracy": 54.12,
-        "bfcl.overall.total_cost_usd": 6.71,
-        "bfcl.overall.latency_mean_s": 5.83,
-        "bfcl.overall.latency_std_s": 11.71,
-        "bfcl.overall.latency_p95_s": 10.59,
-        "bfcl.non_live.ast_accuracy": 34.85,
-        "bfcl.non_live.simple_ast_accuracy": 37.92,
-        "bfcl.non_live.multiple_ast_accuracy": 74.0,
-        "bfcl.non_live.parallel_ast_accuracy": 15.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 12.5,
-        "bfcl.live.live_accuracy": 53.66,
-        "bfcl.live.live_simple_ast_accuracy": 66.28,
-        "bfcl.live.live_multiple_ast_accuracy": 51.66,
-        "bfcl.live.live_parallel_ast_accuracy": 25.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 25.0,
-        "bfcl.multi_turn.accuracy": 37.38,
-        "bfcl.multi_turn.base_accuracy": 41.5,
-        "bfcl.multi_turn.miss_function_accuracy": 39.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 33.5,
-        "bfcl.multi_turn.long_context_accuracy": 35.0,
-        "bfcl.web_search.accuracy": 69.5,
-        "bfcl.web_search.base_accuracy": 80.0,
-        "bfcl.web_search.no_snippet_accuracy": 59.0,
-        "bfcl.memory.accuracy": 54.19,
-        "bfcl.memory.kv_accuracy": 41.94,
-        "bfcl.memory.vector_accuracy": 61.29,
-        "bfcl.memory.recursive_summarization_accuracy": 59.35,
-        "bfcl.relevance.relevance_detection_accuracy": 37.5,
-        "bfcl.relevance.irrelevance_detection_accuracy": 93.18
-      }
-    },
-    {
-      "model_id": "deepseek/deepseek-v3-2-exp-prompt-thinking",
-      "name": "DeepSeek-V3.2-Exp (Prompt + Thinking)",
-      "developer": "deepseek",
-      "scores": {
-        "bfcl.overall.rank": 14.0,
-        "bfcl.overall.overall_accuracy": 56.73,
-        "bfcl.overall.total_cost_usd": 57.75,
-        "bfcl.overall.latency_mean_s": 37.89,
-        "bfcl.overall.latency_std_s": 49.56,
-        "bfcl.overall.latency_p95_s": 102.09,
-        "bfcl.non_live.ast_accuracy": 85.52,
-        "bfcl.non_live.simple_ast_accuracy": 74.08,
-        "bfcl.non_live.multiple_ast_accuracy": 92.0,
-        "bfcl.non_live.parallel_ast_accuracy": 89.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 86.5,
-        "bfcl.live.live_accuracy": 76.02,
-        "bfcl.live.live_simple_ast_accuracy": 82.56,
-        "bfcl.live.live_multiple_ast_accuracy": 74.74,
-        "bfcl.live.live_parallel_ast_accuracy": 87.5,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 54.17,
-        "bfcl.multi_turn.accuracy": 44.88,
-        "bfcl.multi_turn.base_accuracy": 55.0,
-        "bfcl.multi_turn.miss_function_accuracy": 49.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 27.0,
-        "bfcl.multi_turn.long_context_accuracy": 48.5,
-        "bfcl.web_search.accuracy": 58.0,
-        "bfcl.web_search.base_accuracy": 64.0,
-        "bfcl.web_search.no_snippet_accuracy": 52.0,
-        "bfcl.memory.accuracy": 44.09,
-        "bfcl.memory.kv_accuracy": 46.45,
-        "bfcl.memory.vector_accuracy": 46.45,
-        "bfcl.memory.recursive_summarization_accuracy": 39.35,
-        "bfcl.relevance.relevance_detection_accuracy": 93.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 67.0,
-        "bfcl.format_sensitivity.max_delta": 10.0,
-        "bfcl.format_sensitivity.stddev": 2.77
-      }
-    },
-    {
-      "model_id": "google/gemini-2-5-flash-fc",
-      "name": "Gemini-2.5-Flash (FC)",
-      "developer": "Google",
-      "scores": {
-        "bfcl.overall.rank": 15.0,
-        "bfcl.overall.overall_accuracy": 56.24,
-        "bfcl.overall.total_cost_usd": 26.36,
-        "bfcl.overall.latency_mean_s": 2.99,
-        "bfcl.overall.latency_std_s": 9.22,
-        "bfcl.overall.latency_p95_s": 5.62,
-        "bfcl.non_live.ast_accuracy": 84.96,
-        "bfcl.non_live.simple_ast_accuracy": 74.33,
-        "bfcl.non_live.multiple_ast_accuracy": 92.0,
-        "bfcl.non_live.parallel_ast_accuracy": 94.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 79.5,
-        "bfcl.live.live_accuracy": 74.39,
-        "bfcl.live.live_simple_ast_accuracy": 85.27,
-        "bfcl.live.live_multiple_ast_accuracy": 71.7,
-        "bfcl.live.live_parallel_ast_accuracy": 81.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 70.83,
-        "bfcl.multi_turn.accuracy": 36.25,
-        "bfcl.multi_turn.base_accuracy": 41.5,
-        "bfcl.multi_turn.miss_function_accuracy": 36.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 32.0,
-        "bfcl.multi_turn.long_context_accuracy": 35.5,
-        "bfcl.web_search.accuracy": 59.0,
-        "bfcl.web_search.base_accuracy": 59.0,
-        "bfcl.web_search.no_snippet_accuracy": 59.0,
-        "bfcl.memory.accuracy": 41.29,
-        "bfcl.memory.kv_accuracy": 19.35,
-        "bfcl.memory.vector_accuracy": 50.32,
-        "bfcl.memory.recursive_summarization_accuracy": 54.19,
-        "bfcl.relevance.relevance_detection_accuracy": 75.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 93.67
-      }
-    },
-    {
-      "model_id": "google/gemini-2-5-flash-lite-fc",
-      "name": "Gemini-2.5-Flash-Lite (FC)",
-      "developer": "Google",
-      "scores": {
-        "bfcl.overall.rank": 52.0,
-        "bfcl.overall.overall_accuracy": 36.87,
-        "bfcl.overall.total_cost_usd": 7.55,
-        "bfcl.overall.latency_mean_s": 1.18,
-        "bfcl.overall.latency_std_s": 8.06,
-        "bfcl.overall.latency_p95_s": 1.67,
-        "bfcl.non_live.ast_accuracy": 86.6,
-        "bfcl.non_live.simple_ast_accuracy": 70.92,
-        "bfcl.non_live.multiple_ast_accuracy": 90.0,
-        "bfcl.non_live.parallel_ast_accuracy": 93.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 92.0,
-        "bfcl.live.live_accuracy": 65.8,
-        "bfcl.live.live_simple_ast_accuracy": 73.26,
-        "bfcl.live.live_multiple_ast_accuracy": 63.82,
-        "bfcl.live.live_parallel_ast_accuracy": 75.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 66.67,
-        "bfcl.multi_turn.accuracy": 13.5,
-        "bfcl.multi_turn.base_accuracy": 20.0,
-        "bfcl.multi_turn.miss_function_accuracy": 1.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 15.0,
-        "bfcl.multi_turn.long_context_accuracy": 17.5,
-        "bfcl.web_search.accuracy": 21.0,
-        "bfcl.web_search.base_accuracy": 26.0,
-        "bfcl.web_search.no_snippet_accuracy": 16.0,
-        "bfcl.memory.accuracy": 20.65,
-        "bfcl.memory.kv_accuracy": 3.87,
-        "bfcl.memory.vector_accuracy": 6.45,
-        "bfcl.memory.recursive_summarization_accuracy": 51.61,
-        "bfcl.relevance.relevance_detection_accuracy": 43.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 92.5
-      }
-    },
-    {
-      "model_id": "google/gemini-2-5-flash-lite-prompt",
-      "name": "Gemini-2.5-Flash-Lite (Prompt)",
-      "developer": "Google",
-      "scores": {
-        "bfcl.overall.rank": 73.0,
-        "bfcl.overall.overall_accuracy": 28.03,
-        "bfcl.overall.total_cost_usd": 7.05,
-        "bfcl.overall.latency_mean_s": 1.0,
-        "bfcl.overall.latency_std_s": 4.75,
-        "bfcl.overall.latency_p95_s": 1.4,
-        "bfcl.non_live.ast_accuracy": 83.9,
-        "bfcl.non_live.simple_ast_accuracy": 70.08,
-        "bfcl.non_live.multiple_ast_accuracy": 86.0,
-        "bfcl.non_live.parallel_ast_accuracy": 90.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 89.5,
-        "bfcl.live.live_accuracy": 54.85,
-        "bfcl.live.live_simple_ast_accuracy": 67.05,
-        "bfcl.live.live_multiple_ast_accuracy": 51.66,
-        "bfcl.live.live_parallel_ast_accuracy": 75.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 50.0,
-        "bfcl.multi_turn.accuracy": 7.63,
-        "bfcl.multi_turn.base_accuracy": 10.0,
-        "bfcl.multi_turn.miss_function_accuracy": 5.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 6.5,
-        "bfcl.multi_turn.long_context_accuracy": 9.0,
-        "bfcl.web_search.accuracy": 0.0,
-        "bfcl.web_search.base_accuracy": 0.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 12.69,
-        "bfcl.memory.kv_accuracy": 1.94,
-        "bfcl.memory.vector_accuracy": 6.45,
-        "bfcl.memory.recursive_summarization_accuracy": 29.68,
-        "bfcl.relevance.relevance_detection_accuracy": 50.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 93.33,
-        "bfcl.format_sensitivity.max_delta": 25.5,
-        "bfcl.format_sensitivity.stddev": 6.68
-      }
-    },
-    {
-      "model_id": "google/gemini-2-5-flash-prompt",
-      "name": "Gemini-2.5-Flash (Prompt)",
-      "developer": "Google",
-      "scores": {
-        "bfcl.overall.rank": 26.0,
-        "bfcl.overall.overall_accuracy": 50.9,
-        "bfcl.overall.total_cost_usd": 33.45,
-        "bfcl.overall.latency_mean_s": 3.18,
-        "bfcl.overall.latency_std_s": 4.44,
-        "bfcl.overall.latency_p95_s": 6.09,
-        "bfcl.non_live.ast_accuracy": 88.08,
-        "bfcl.non_live.simple_ast_accuracy": 77.33,
-        "bfcl.non_live.multiple_ast_accuracy": 91.5,
-        "bfcl.non_live.parallel_ast_accuracy": 96.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 87.5,
-        "bfcl.live.live_accuracy": 78.16,
-        "bfcl.live.live_simple_ast_accuracy": 87.21,
-        "bfcl.live.live_multiple_ast_accuracy": 75.97,
-        "bfcl.live.live_parallel_ast_accuracy": 81.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 75.0,
-        "bfcl.multi_turn.accuracy": 16.75,
-        "bfcl.multi_turn.base_accuracy": 14.5,
-        "bfcl.multi_turn.miss_function_accuracy": 16.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 17.5,
-        "bfcl.multi_turn.long_context_accuracy": 18.5,
-        "bfcl.web_search.accuracy": 62.0,
-        "bfcl.web_search.base_accuracy": 60.0,
-        "bfcl.web_search.no_snippet_accuracy": 64.0,
-        "bfcl.memory.accuracy": 38.71,
-        "bfcl.memory.kv_accuracy": 13.55,
-        "bfcl.memory.vector_accuracy": 47.1,
-        "bfcl.memory.recursive_summarization_accuracy": 55.48,
-        "bfcl.relevance.relevance_detection_accuracy": 62.5,
-        "bfcl.relevance.irrelevance_detection_accuracy": 91.09,
-        "bfcl.format_sensitivity.max_delta": 9.0,
-        "bfcl.format_sensitivity.stddev": 2.45
-      }
-    },
-    {
-      "model_id": "google/gemini-3-pro-preview-fc",
-      "name": "Gemini-3-Pro-Preview (FC)",
-      "developer": "Google",
-      "scores": {
-        "bfcl.overall.rank": 7.0,
-        "bfcl.overall.overall_accuracy": 68.14,
-        "bfcl.overall.total_cost_usd": 224.69,
-        "bfcl.overall.latency_mean_s": 15.87,
-        "bfcl.overall.latency_std_s": 41.41,
-        "bfcl.overall.latency_p95_s": 58.48,
-        "bfcl.non_live.ast_accuracy": 85.75,
-        "bfcl.non_live.simple_ast_accuracy": 75.5,
-        "bfcl.non_live.multiple_ast_accuracy": 94.0,
-        "bfcl.non_live.parallel_ast_accuracy": 91.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 82.5,
-        "bfcl.live.live_accuracy": 81.72,
-        "bfcl.live.live_simple_ast_accuracy": 87.6,
-        "bfcl.live.live_multiple_ast_accuracy": 80.44,
-        "bfcl.live.live_parallel_ast_accuracy": 75.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 79.17,
-        "bfcl.multi_turn.accuracy": 63.12,
-        "bfcl.multi_turn.base_accuracy": 69.0,
-        "bfcl.multi_turn.miss_function_accuracy": 63.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 56.5,
-        "bfcl.multi_turn.long_context_accuracy": 64.0,
-        "bfcl.web_search.accuracy": 68.5,
-        "bfcl.web_search.base_accuracy": 63.0,
-        "bfcl.web_search.no_snippet_accuracy": 74.0,
-        "bfcl.memory.accuracy": 54.84,
-        "bfcl.memory.kv_accuracy": 50.32,
-        "bfcl.memory.vector_accuracy": 63.23,
-        "bfcl.memory.recursive_summarization_accuracy": 50.97,
-        "bfcl.relevance.relevance_detection_accuracy": 75.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 77.85
-      }
-    },
-    {
-      "model_id": "google/gemini-3-pro-preview-prompt",
-      "name": "Gemini-3-Pro-Preview (Prompt)",
-      "developer": "Google",
-      "scores": {
-        "bfcl.overall.rank": 3.0,
-        "bfcl.overall.overall_accuracy": 72.51,
-        "bfcl.overall.total_cost_usd": 298.47,
-        "bfcl.overall.latency_mean_s": 12.08,
-        "bfcl.overall.latency_std_s": 21.3,
-        "bfcl.overall.latency_p95_s": 32.73,
-        "bfcl.non_live.ast_accuracy": 90.65,
-        "bfcl.non_live.simple_ast_accuracy": 79.58,
-        "bfcl.non_live.multiple_ast_accuracy": 96.0,
-        "bfcl.non_live.parallel_ast_accuracy": 95.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 92.0,
-        "bfcl.live.live_accuracy": 83.12,
-        "bfcl.live.live_simple_ast_accuracy": 87.6,
-        "bfcl.live.live_multiple_ast_accuracy": 81.77,
-        "bfcl.live.live_parallel_ast_accuracy": 93.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 87.5,
-        "bfcl.multi_turn.accuracy": 60.75,
-        "bfcl.multi_turn.base_accuracy": 64.5,
-        "bfcl.multi_turn.miss_function_accuracy": 60.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 54.5,
-        "bfcl.multi_turn.long_context_accuracy": 64.0,
-        "bfcl.web_search.accuracy": 80.0,
-        "bfcl.web_search.base_accuracy": 78.0,
-        "bfcl.web_search.no_snippet_accuracy": 82.0,
-        "bfcl.memory.accuracy": 61.72,
-        "bfcl.memory.kv_accuracy": 59.35,
-        "bfcl.memory.vector_accuracy": 62.58,
-        "bfcl.memory.recursive_summarization_accuracy": 63.23,
-        "bfcl.relevance.relevance_detection_accuracy": 68.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 85.59,
-        "bfcl.format_sensitivity.max_delta": 8.5,
-        "bfcl.format_sensitivity.stddev": 1.7
-      }
-    },
-    {
-      "model_id": "google/gemma-3-12b-it-prompt",
-      "name": "Gemma-3-12b-it (Prompt)",
-      "developer": "Google",
-      "scores": {
-        "bfcl.overall.rank": 66.0,
-        "bfcl.overall.overall_accuracy": 30.43,
-        "bfcl.overall.total_cost_usd": 10.77,
-        "bfcl.overall.latency_mean_s": 11.1,
-        "bfcl.overall.latency_std_s": 17.17,
-        "bfcl.overall.latency_p95_s": 34.66,
-        "bfcl.non_live.ast_accuracy": 79.44,
-        "bfcl.non_live.simple_ast_accuracy": 76.25,
-        "bfcl.non_live.multiple_ast_accuracy": 94.0,
-        "bfcl.non_live.parallel_ast_accuracy": 91.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 56.5,
-        "bfcl.live.live_accuracy": 74.24,
-        "bfcl.live.live_simple_ast_accuracy": 85.66,
-        "bfcl.live.live_multiple_ast_accuracy": 71.89,
-        "bfcl.live.live_parallel_ast_accuracy": 87.5,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 45.83,
-        "bfcl.multi_turn.accuracy": 5.75,
-        "bfcl.multi_turn.base_accuracy": 6.5,
-        "bfcl.multi_turn.miss_function_accuracy": 7.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 5.0,
-        "bfcl.multi_turn.long_context_accuracy": 4.0,
-        "bfcl.web_search.accuracy": 4.0,
-        "bfcl.web_search.base_accuracy": 4.0,
-        "bfcl.web_search.no_snippet_accuracy": 4.0,
-        "bfcl.memory.accuracy": 27.53,
-        "bfcl.memory.kv_accuracy": 8.39,
-        "bfcl.memory.vector_accuracy": 25.16,
-        "bfcl.memory.recursive_summarization_accuracy": 49.03,
-        "bfcl.relevance.relevance_detection_accuracy": 93.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 70.29,
-        "bfcl.format_sensitivity.max_delta": 67.5,
-        "bfcl.format_sensitivity.stddev": 22.41
-      }
-    },
-    {
-      "model_id": "google/gemma-3-1b-it-prompt",
-      "name": "Gemma-3-1b-it (Prompt)",
-      "developer": "Google",
-      "scores": {
-        "bfcl.overall.rank": 109.0,
-        "bfcl.overall.overall_accuracy": 7.17,
-        "bfcl.overall.total_cost_usd": 3.4,
-        "bfcl.overall.latency_mean_s": 3.98,
-        "bfcl.overall.latency_std_s": 9.8,
-        "bfcl.overall.latency_p95_s": 12.06,
-        "bfcl.non_live.ast_accuracy": 20.21,
-        "bfcl.non_live.simple_ast_accuracy": 43.33,
-        "bfcl.non_live.multiple_ast_accuracy": 36.0,
-        "bfcl.non_live.parallel_ast_accuracy": 0.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 1.5,
-        "bfcl.live.live_accuracy": 11.84,
-        "bfcl.live.live_simple_ast_accuracy": 36.43,
-        "bfcl.live.live_multiple_ast_accuracy": 6.27,
-        "bfcl.live.live_parallel_ast_accuracy": 0.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 0.0,
-        "bfcl.multi_turn.accuracy": 0.0,
-        "bfcl.multi_turn.base_accuracy": 0.0,
-        "bfcl.multi_turn.miss_function_accuracy": 0.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 0.0,
-        "bfcl.multi_turn.long_context_accuracy": 0.0,
-        "bfcl.web_search.accuracy": 0.0,
-        "bfcl.web_search.base_accuracy": 0.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 3.23,
-        "bfcl.memory.kv_accuracy": 3.87,
-        "bfcl.memory.vector_accuracy": 3.87,
-        "bfcl.memory.recursive_summarization_accuracy": 1.94,
-        "bfcl.relevance.relevance_detection_accuracy": 37.5,
-        "bfcl.relevance.irrelevance_detection_accuracy": 33.18,
-        "bfcl.format_sensitivity.max_delta": 25.5,
-        "bfcl.format_sensitivity.stddev": 9.76
-      }
-    },
-    {
-      "model_id": "google/gemma-3-27b-it-prompt",
-      "name": "Gemma-3-27b-it (Prompt)",
-      "developer": "Google",
-      "scores": {
-        "bfcl.overall.rank": 69.0,
-        "bfcl.overall.overall_accuracy": 29.47,
-        "bfcl.overall.total_cost_usd": 11.82,
-        "bfcl.overall.latency_mean_s": 10.88,
-        "bfcl.overall.latency_std_s": 19.67,
-        "bfcl.overall.latency_p95_s": 55.5,
-        "bfcl.non_live.ast_accuracy": 87.17,
-        "bfcl.non_live.simple_ast_accuracy": 77.67,
-        "bfcl.non_live.multiple_ast_accuracy": 92.5,
-        "bfcl.non_live.parallel_ast_accuracy": 89.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 89.5,
-        "bfcl.live.live_accuracy": 74.54,
-        "bfcl.live.live_simple_ast_accuracy": 84.5,
-        "bfcl.live.live_multiple_ast_accuracy": 72.46,
-        "bfcl.live.live_parallel_ast_accuracy": 93.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 45.83,
-        "bfcl.multi_turn.accuracy": 10.75,
-        "bfcl.multi_turn.base_accuracy": 16.5,
-        "bfcl.multi_turn.miss_function_accuracy": 4.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 8.0,
-        "bfcl.multi_turn.long_context_accuracy": 14.0,
-        "bfcl.web_search.accuracy": 0.0,
-        "bfcl.web_search.base_accuracy": 0.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 13.55,
-        "bfcl.memory.kv_accuracy": 1.94,
-        "bfcl.memory.vector_accuracy": 3.23,
-        "bfcl.memory.recursive_summarization_accuracy": 35.48,
-        "bfcl.relevance.relevance_detection_accuracy": 81.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 73.67,
-        "bfcl.format_sensitivity.max_delta": 34.0,
-        "bfcl.format_sensitivity.stddev": 8.06
-      }
-    },
-    {
-      "model_id": "google/gemma-3-4b-it-prompt",
-      "name": "Gemma-3-4b-it (Prompt)",
-      "developer": "Google",
-      "scores": {
-        "bfcl.overall.rank": 101.0,
-        "bfcl.overall.overall_accuracy": 19.62,
-        "bfcl.overall.total_cost_usd": 4.14,
-        "bfcl.overall.latency_mean_s": 4.69,
-        "bfcl.overall.latency_std_s": 9.53,
-        "bfcl.overall.latency_p95_s": 11.42,
-        "bfcl.non_live.ast_accuracy": 61.12,
-        "bfcl.non_live.simple_ast_accuracy": 64.5,
-        "bfcl.non_live.multiple_ast_accuracy": 88.0,
-        "bfcl.non_live.parallel_ast_accuracy": 56.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 36.0,
-        "bfcl.live.live_accuracy": 60.84,
-        "bfcl.live.live_simple_ast_accuracy": 70.93,
-        "bfcl.live.live_multiple_ast_accuracy": 59.35,
-        "bfcl.live.live_parallel_ast_accuracy": 25.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 41.67,
-        "bfcl.multi_turn.accuracy": 0.38,
-        "bfcl.multi_turn.base_accuracy": 0.5,
-        "bfcl.multi_turn.miss_function_accuracy": 0.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 0.5,
-        "bfcl.multi_turn.long_context_accuracy": 0.5,
-        "bfcl.web_search.accuracy": 1.0,
-        "bfcl.web_search.base_accuracy": 1.0,
-        "bfcl.web_search.no_snippet_accuracy": 1.0,
-        "bfcl.memory.accuracy": 8.6,
-        "bfcl.memory.kv_accuracy": 9.68,
-        "bfcl.memory.vector_accuracy": 9.68,
-        "bfcl.memory.recursive_summarization_accuracy": 6.45,
-        "bfcl.relevance.relevance_detection_accuracy": 100.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 53.94,
-        "bfcl.format_sensitivity.max_delta": 69.5,
-        "bfcl.format_sensitivity.stddev": 23.67
-      }
-    },
-    {
-      "model_id": "huawei-noah-ustc/toolace-2-8b-fc",
-      "name": "ToolACE-2-8B (FC)",
-      "developer": "huawei-noah-ustc",
-      "scores": {
-        "bfcl.overall.rank": 40.0,
-        "bfcl.overall.overall_accuracy": 42.44,
-        "bfcl.overall.total_cost_usd": 24.43,
-        "bfcl.overall.latency_mean_s": 15.95,
-        "bfcl.overall.latency_std_s": 40.06,
-        "bfcl.overall.latency_p95_s": 65.26,
-        "bfcl.non_live.ast_accuracy": 87.1,
-        "bfcl.non_live.simple_ast_accuracy": 73.42,
-        "bfcl.non_live.multiple_ast_accuracy": 91.0,
-        "bfcl.non_live.parallel_ast_accuracy": 93.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 91.0,
-        "bfcl.live.live_accuracy": 77.42,
-        "bfcl.live.live_simple_ast_accuracy": 71.32,
-        "bfcl.live.live_multiple_ast_accuracy": 79.39,
-        "bfcl.live.live_parallel_ast_accuracy": 68.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 62.5,
-        "bfcl.multi_turn.accuracy": 38.38,
-        "bfcl.multi_turn.base_accuracy": 49.0,
-        "bfcl.multi_turn.miss_function_accuracy": 28.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 30.5,
-        "bfcl.multi_turn.long_context_accuracy": 46.0,
-        "bfcl.web_search.accuracy": 8.5,
-        "bfcl.web_search.base_accuracy": 13.0,
-        "bfcl.web_search.no_snippet_accuracy": 4.0,
-        "bfcl.memory.accuracy": 18.49,
-        "bfcl.memory.kv_accuracy": 5.81,
-        "bfcl.memory.vector_accuracy": 16.13,
-        "bfcl.memory.recursive_summarization_accuracy": 33.55,
-        "bfcl.relevance.relevance_detection_accuracy": 75.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 90.79,
-        "bfcl.format_sensitivity.max_delta": 81.5,
-        "bfcl.format_sensitivity.stddev": 27.92
-      }
-    },
-    {
-      "model_id": "ibm/granite-20b-functioncalling-fc",
-      "name": "Granite-20b-FunctionCalling (FC)",
-      "developer": "ibm",
-      "scores": {
-        "bfcl.overall.rank": 93.0,
-        "bfcl.overall.overall_accuracy": 23.23,
-        "bfcl.overall.total_cost_usd": 5.23,
-        "bfcl.overall.latency_mean_s": 3.2,
-        "bfcl.overall.latency_std_s": 3.43,
-        "bfcl.overall.latency_p95_s": 9.97,
-        "bfcl.non_live.ast_accuracy": 82.35,
-        "bfcl.non_live.simple_ast_accuracy": 72.92,
-        "bfcl.non_live.multiple_ast_accuracy": 91.5,
-        "bfcl.non_live.parallel_ast_accuracy": 83.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 81.5,
-        "bfcl.live.live_accuracy": 58.7,
-        "bfcl.live.live_simple_ast_accuracy": 67.83,
-        "bfcl.live.live_multiple_ast_accuracy": 56.7,
-        "bfcl.live.live_parallel_ast_accuracy": 43.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 58.33,
-        "bfcl.multi_turn.accuracy": 5.38,
-        "bfcl.multi_turn.base_accuracy": 9.0,
-        "bfcl.multi_turn.miss_function_accuracy": 3.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 6.5,
-        "bfcl.multi_turn.long_context_accuracy": 3.0,
-        "bfcl.web_search.accuracy": 0.0,
-        "bfcl.web_search.base_accuracy": 0.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 0.0,
-        "bfcl.memory.kv_accuracy": 0.0,
-        "bfcl.memory.vector_accuracy": 0.0,
-        "bfcl.memory.recursive_summarization_accuracy": 0.0,
-        "bfcl.relevance.relevance_detection_accuracy": 87.5,
-        "bfcl.relevance.irrelevance_detection_accuracy": 75.13
-      }
-    },
-    {
-      "model_id": "ibm/granite-3-1-8b-instruct-fc",
-      "name": "Granite-3.1-8B-Instruct (FC)",
-      "developer": "ibm",
-      "scores": {
-        "bfcl.overall.rank": 81.0,
-        "bfcl.overall.overall_accuracy": 27.1,
-        "bfcl.overall.total_cost_usd": 9.32,
-        "bfcl.overall.latency_mean_s": 13.23,
-        "bfcl.overall.latency_std_s": 31.28,
-        "bfcl.overall.latency_p95_s": 65.19,
-        "bfcl.non_live.ast_accuracy": 78.33,
-        "bfcl.non_live.simple_ast_accuracy": 67.33,
-        "bfcl.non_live.multiple_ast_accuracy": 92.0,
-        "bfcl.non_live.parallel_ast_accuracy": 84.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 70.0,
-        "bfcl.live.live_accuracy": 60.33,
-        "bfcl.live.live_simple_ast_accuracy": 58.53,
-        "bfcl.live.live_multiple_ast_accuracy": 61.82,
-        "bfcl.live.live_parallel_ast_accuracy": 18.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 41.67,
-        "bfcl.multi_turn.accuracy": 7.5,
-        "bfcl.multi_turn.base_accuracy": 11.5,
-        "bfcl.multi_turn.miss_function_accuracy": 2.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 7.5,
-        "bfcl.multi_turn.long_context_accuracy": 9.0,
-        "bfcl.web_search.accuracy": 0.5,
-        "bfcl.web_search.base_accuracy": 1.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 14.41,
-        "bfcl.memory.kv_accuracy": 9.68,
-        "bfcl.memory.vector_accuracy": 7.1,
-        "bfcl.memory.recursive_summarization_accuracy": 26.45,
-        "bfcl.relevance.relevance_detection_accuracy": 68.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 79.98
-      }
-    },
-    {
-      "model_id": "ibm/granite-3-2-8b-instruct-fc",
-      "name": "Granite-3.2-8B-Instruct (FC)",
-      "developer": "ibm",
-      "scores": {
-        "bfcl.overall.rank": 83.0,
-        "bfcl.overall.overall_accuracy": 26.87,
-        "bfcl.overall.total_cost_usd": 25.02,
-        "bfcl.overall.latency_mean_s": 36.13,
-        "bfcl.overall.latency_std_s": 81.76,
-        "bfcl.overall.latency_p95_s": 216.28,
-        "bfcl.non_live.ast_accuracy": 79.77,
-        "bfcl.non_live.simple_ast_accuracy": 69.58,
-        "bfcl.non_live.multiple_ast_accuracy": 88.5,
-        "bfcl.non_live.parallel_ast_accuracy": 88.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 72.5,
-        "bfcl.live.live_accuracy": 60.33,
-        "bfcl.live.live_simple_ast_accuracy": 60.47,
-        "bfcl.live.live_multiple_ast_accuracy": 61.16,
-        "bfcl.live.live_parallel_ast_accuracy": 25.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 45.83,
-        "bfcl.multi_turn.accuracy": 7.38,
-        "bfcl.multi_turn.base_accuracy": 9.5,
-        "bfcl.multi_turn.miss_function_accuracy": 3.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 8.0,
-        "bfcl.multi_turn.long_context_accuracy": 9.0,
-        "bfcl.web_search.accuracy": 0.5,
-        "bfcl.web_search.base_accuracy": 1.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 12.47,
-        "bfcl.memory.kv_accuracy": 6.45,
-        "bfcl.memory.vector_accuracy": 9.68,
-        "bfcl.memory.recursive_summarization_accuracy": 21.29,
-        "bfcl.relevance.relevance_detection_accuracy": 75.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 80.53
-      }
-    },
-    {
-      "model_id": "ibm/granite-4-0-350m-fc",
-      "name": "Granite-4.0-350m (FC)",
-      "developer": "ibm",
-      "scores": {
-        "bfcl.overall.rank": 103.0,
-        "bfcl.overall.overall_accuracy": 18.98,
-        "bfcl.overall.total_cost_usd": 1.44,
-        "bfcl.overall.latency_mean_s": 1.74,
-        "bfcl.overall.latency_std_s": 4.85,
-        "bfcl.overall.latency_p95_s": 3.44,
-        "bfcl.non_live.ast_accuracy": 67.92,
-        "bfcl.non_live.simple_ast_accuracy": 61.67,
-        "bfcl.non_live.multiple_ast_accuracy": 84.5,
-        "bfcl.non_live.parallel_ast_accuracy": 70.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 55.5,
-        "bfcl.live.live_accuracy": 46.11,
-        "bfcl.live.live_simple_ast_accuracy": 61.24,
-        "bfcl.live.live_multiple_ast_accuracy": 42.36,
-        "bfcl.live.live_parallel_ast_accuracy": 68.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 33.33,
-        "bfcl.multi_turn.accuracy": 2.5,
-        "bfcl.multi_turn.base_accuracy": 5.0,
-        "bfcl.multi_turn.miss_function_accuracy": 0.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 2.5,
-        "bfcl.multi_turn.long_context_accuracy": 2.0,
-        "bfcl.web_search.accuracy": 0.5,
-        "bfcl.web_search.base_accuracy": 0.0,
-        "bfcl.web_search.no_snippet_accuracy": 1.0,
-        "bfcl.memory.accuracy": 3.23,
-        "bfcl.memory.kv_accuracy": 1.94,
-        "bfcl.memory.vector_accuracy": 1.29,
-        "bfcl.memory.recursive_summarization_accuracy": 6.45,
-        "bfcl.relevance.relevance_detection_accuracy": 81.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 60.84
-      }
-    },
-    {
-      "model_id": "katanemo/arch-agent-1-5b",
-      "name": "Arch-Agent-1.5B",
-      "developer": "katanemo",
-      "scores": {
-        "bfcl.overall.rank": 60.0,
-        "bfcl.overall.overall_accuracy": 32.14,
-        "bfcl.overall.total_cost_usd": 2.45,
-        "bfcl.overall.latency_mean_s": 2.38,
-        "bfcl.overall.latency_std_s": 4.01,
-        "bfcl.overall.latency_p95_s": 5.3,
-        "bfcl.non_live.ast_accuracy": 82.67,
-        "bfcl.non_live.simple_ast_accuracy": 72.17,
-        "bfcl.non_live.multiple_ast_accuracy": 92.0,
-        "bfcl.non_live.parallel_ast_accuracy": 85.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 81.0,
-        "bfcl.live.live_accuracy": 67.73,
-        "bfcl.live.live_simple_ast_accuracy": 70.54,
-        "bfcl.live.live_multiple_ast_accuracy": 67.81,
-        "bfcl.live.live_parallel_ast_accuracy": 31.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 58.33,
-        "bfcl.multi_turn.accuracy": 26.62,
-        "bfcl.multi_turn.base_accuracy": 35.5,
-        "bfcl.multi_turn.miss_function_accuracy": 27.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 21.5,
-        "bfcl.multi_turn.long_context_accuracy": 22.0,
-        "bfcl.web_search.accuracy": 0.0,
-        "bfcl.web_search.base_accuracy": 0.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 8.17,
-        "bfcl.memory.kv_accuracy": 5.81,
-        "bfcl.memory.vector_accuracy": 5.81,
-        "bfcl.memory.recursive_summarization_accuracy": 12.9,
-        "bfcl.relevance.relevance_detection_accuracy": 75.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 74.83
-      }
-    },
-    {
-      "model_id": "katanemo/arch-agent-32b",
-      "name": "Arch-Agent-32B",
-      "developer": "katanemo",
-      "scores": {
-        "bfcl.overall.rank": 37.0,
-        "bfcl.overall.overall_accuracy": 45.37,
-        "bfcl.overall.total_cost_usd": 8.87,
-        "bfcl.overall.latency_mean_s": 9.44,
-        "bfcl.overall.latency_std_s": 21.44,
-        "bfcl.overall.latency_p95_s": 24.87,
-        "bfcl.non_live.ast_accuracy": 88.92,
-        "bfcl.non_live.simple_ast_accuracy": 76.67,
-        "bfcl.non_live.multiple_ast_accuracy": 94.0,
-        "bfcl.non_live.parallel_ast_accuracy": 96.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 89.0,
-        "bfcl.live.live_accuracy": 80.68,
-        "bfcl.live.live_simple_ast_accuracy": 86.43,
-        "bfcl.live.live_multiple_ast_accuracy": 79.11,
-        "bfcl.live.live_parallel_ast_accuracy": 93.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 79.17,
-        "bfcl.multi_turn.accuracy": 54.25,
-        "bfcl.multi_turn.base_accuracy": 64.5,
-        "bfcl.multi_turn.miss_function_accuracy": 58.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 53.0,
-        "bfcl.multi_turn.long_context_accuracy": 41.5,
-        "bfcl.web_search.accuracy": 5.0,
-        "bfcl.web_search.base_accuracy": 4.0,
-        "bfcl.web_search.no_snippet_accuracy": 6.0,
-        "bfcl.memory.accuracy": 14.62,
-        "bfcl.memory.kv_accuracy": 5.81,
-        "bfcl.memory.vector_accuracy": 9.03,
-        "bfcl.memory.recursive_summarization_accuracy": 29.03,
-        "bfcl.relevance.relevance_detection_accuracy": 81.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 82.15
-      }
-    },
-    {
-      "model_id": "katanemo/arch-agent-3b",
-      "name": "Arch-Agent-3B",
-      "developer": "katanemo",
-      "scores": {
-        "bfcl.overall.rank": 56.0,
-        "bfcl.overall.overall_accuracy": 35.36,
-        "bfcl.overall.total_cost_usd": 3.7,
-        "bfcl.overall.latency_mean_s": 3.56,
-        "bfcl.overall.latency_std_s": 6.65,
-        "bfcl.overall.latency_p95_s": 8.19,
-        "bfcl.non_live.ast_accuracy": 86.67,
-        "bfcl.non_live.simple_ast_accuracy": 78.67,
-        "bfcl.non_live.multiple_ast_accuracy": 94.5,
-        "bfcl.non_live.parallel_ast_accuracy": 91.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 82.5,
-        "bfcl.live.live_accuracy": 72.91,
-        "bfcl.live.live_simple_ast_accuracy": 75.58,
-        "bfcl.live.live_multiple_ast_accuracy": 72.27,
-        "bfcl.live.live_parallel_ast_accuracy": 68.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 75.0,
-        "bfcl.multi_turn.accuracy": 34.88,
-        "bfcl.multi_turn.base_accuracy": 42.0,
-        "bfcl.multi_turn.miss_function_accuracy": 37.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 31.0,
-        "bfcl.multi_turn.long_context_accuracy": 29.0,
-        "bfcl.web_search.accuracy": 0.5,
-        "bfcl.web_search.base_accuracy": 1.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 6.88,
-        "bfcl.memory.kv_accuracy": 5.16,
-        "bfcl.memory.vector_accuracy": 5.81,
-        "bfcl.memory.recursive_summarization_accuracy": 9.68,
-        "bfcl.relevance.relevance_detection_accuracy": 68.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 74.67
-      }
-    },
-    {
-      "model_id": "madeagents/hammer2-1-0-5b-fc",
-      "name": "Hammer2.1-0.5b (FC)",
-      "developer": "madeagents",
-      "scores": {
-        "bfcl.overall.rank": 100.0,
-        "bfcl.overall.overall_accuracy": 21.22,
-        "bfcl.overall.total_cost_usd": 2.82,
-        "bfcl.overall.latency_mean_s": 2.79,
-        "bfcl.overall.latency_std_s": 3.17,
-        "bfcl.overall.latency_p95_s": 9.86,
-        "bfcl.non_live.ast_accuracy": 65.98,
-        "bfcl.non_live.simple_ast_accuracy": 62.42,
-        "bfcl.non_live.multiple_ast_accuracy": 81.0,
-        "bfcl.non_live.parallel_ast_accuracy": 69.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 51.5,
-        "bfcl.live.live_accuracy": 54.63,
-        "bfcl.live.live_simple_ast_accuracy": 56.59,
-        "bfcl.live.live_multiple_ast_accuracy": 54.42,
-        "bfcl.live.live_parallel_ast_accuracy": 62.5,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 37.5,
-        "bfcl.multi_turn.accuracy": 2.88,
-        "bfcl.multi_turn.base_accuracy": 4.5,
-        "bfcl.multi_turn.miss_function_accuracy": 0.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 4.0,
-        "bfcl.multi_turn.long_context_accuracy": 2.5,
-        "bfcl.web_search.accuracy": 0.0,
-        "bfcl.web_search.base_accuracy": 0.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 1.08,
-        "bfcl.memory.kv_accuracy": 0.65,
-        "bfcl.memory.vector_accuracy": 1.94,
-        "bfcl.memory.recursive_summarization_accuracy": 0.65,
-        "bfcl.relevance.relevance_detection_accuracy": 68.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 80.79
-      }
-    },
-    {
-      "model_id": "madeagents/hammer2-1-1-5b-fc",
-      "name": "Hammer2.1-1.5b (FC)",
-      "developer": "madeagents",
-      "scores": {
-        "bfcl.overall.rank": 75.0,
-        "bfcl.overall.overall_accuracy": 27.88,
-        "bfcl.overall.total_cost_usd": 6.83,
-        "bfcl.overall.latency_mean_s": 6.28,
-        "bfcl.overall.latency_std_s": 8.79,
-        "bfcl.overall.latency_p95_s": 30.72,
-        "bfcl.non_live.ast_accuracy": 82.98,
-        "bfcl.non_live.simple_ast_accuracy": 73.92,
-        "bfcl.non_live.multiple_ast_accuracy": 92.0,
-        "bfcl.non_live.parallel_ast_accuracy": 85.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 80.5,
-        "bfcl.live.live_accuracy": 69.5,
-        "bfcl.live.live_simple_ast_accuracy": 72.09,
-        "bfcl.live.live_multiple_ast_accuracy": 69.33,
-        "bfcl.live.live_parallel_ast_accuracy": 50.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 62.5,
-        "bfcl.multi_turn.accuracy": 15.62,
-        "bfcl.multi_turn.base_accuracy": 20.0,
-        "bfcl.multi_turn.miss_function_accuracy": 16.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 9.5,
-        "bfcl.multi_turn.long_context_accuracy": 16.5,
-        "bfcl.web_search.accuracy": 0.0,
-        "bfcl.web_search.base_accuracy": 0.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 0.0,
-        "bfcl.memory.kv_accuracy": 0.0,
-        "bfcl.memory.vector_accuracy": 0.0,
-        "bfcl.memory.recursive_summarization_accuracy": 0.0,
-        "bfcl.relevance.relevance_detection_accuracy": 75.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 79.4
-      }
-    },
-    {
-      "model_id": "madeagents/hammer2-1-3b-fc",
-      "name": "Hammer2.1-3b (FC)",
-      "developer": "madeagents",
-      "scores": {
-        "bfcl.overall.rank": 68.0,
-        "bfcl.overall.overall_accuracy": 29.71,
-        "bfcl.overall.total_cost_usd": 10.89,
-        "bfcl.overall.latency_mean_s": 11.24,
-        "bfcl.overall.latency_std_s": 15.81,
-        "bfcl.overall.latency_p95_s": 47.44,
-        "bfcl.non_live.ast_accuracy": 84.96,
-        "bfcl.non_live.simple_ast_accuracy": 79.33,
-        "bfcl.non_live.multiple_ast_accuracy": 93.5,
-        "bfcl.non_live.parallel_ast_accuracy": 86.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 80.5,
-        "bfcl.live.live_accuracy": 70.54,
-        "bfcl.live.live_simple_ast_accuracy": 68.22,
-        "bfcl.live.live_multiple_ast_accuracy": 71.32,
-        "bfcl.live.live_parallel_ast_accuracy": 62.5,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 66.67,
-        "bfcl.multi_turn.accuracy": 16.5,
-        "bfcl.multi_turn.base_accuracy": 22.0,
-        "bfcl.multi_turn.miss_function_accuracy": 12.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 16.0,
-        "bfcl.multi_turn.long_context_accuracy": 15.5,
-        "bfcl.web_search.accuracy": 0.0,
-        "bfcl.web_search.base_accuracy": 0.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 3.01,
-        "bfcl.memory.kv_accuracy": 2.58,
-        "bfcl.memory.vector_accuracy": 3.87,
-        "bfcl.memory.recursive_summarization_accuracy": 2.58,
-        "bfcl.relevance.relevance_detection_accuracy": 56.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 86.12
-      }
-    },
-    {
-      "model_id": "madeagents/hammer2-1-7b-fc",
-      "name": "Hammer2.1-7b (FC)",
-      "developer": "madeagents",
-      "scores": {
-        "bfcl.overall.rank": 64.0,
-        "bfcl.overall.overall_accuracy": 31.67,
-        "bfcl.overall.total_cost_usd": 4.99,
-        "bfcl.overall.latency_mean_s": 5.77,
-        "bfcl.overall.latency_std_s": 10.29,
-        "bfcl.overall.latency_p95_s": 31.26,
-        "bfcl.non_live.ast_accuracy": 85.5,
-        "bfcl.non_live.simple_ast_accuracy": 72.5,
-        "bfcl.non_live.multiple_ast_accuracy": 92.5,
-        "bfcl.non_live.parallel_ast_accuracy": 91.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 86.0,
-        "bfcl.live.live_accuracy": 69.5,
-        "bfcl.live.live_simple_ast_accuracy": 66.67,
-        "bfcl.live.live_multiple_ast_accuracy": 69.99,
-        "bfcl.live.live_parallel_ast_accuracy": 75.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 75.0,
-        "bfcl.multi_turn.accuracy": 23.87,
-        "bfcl.multi_turn.base_accuracy": 24.0,
-        "bfcl.multi_turn.miss_function_accuracy": 28.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 21.5,
-        "bfcl.multi_turn.long_context_accuracy": 21.5,
-        "bfcl.web_search.accuracy": 0.0,
-        "bfcl.web_search.base_accuracy": 0.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 0.0,
-        "bfcl.memory.kv_accuracy": 0.0,
-        "bfcl.memory.vector_accuracy": 0.0,
-        "bfcl.memory.recursive_summarization_accuracy": 0.0,
-        "bfcl.relevance.relevance_detection_accuracy": 50.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 90.12
-      }
-    },
-    {
-      "model_id": "meta/llama-3-1-8b-instruct-prompt",
-      "name": "Llama-3.1-8B-Instruct (Prompt)",
-      "developer": "Meta",
-      "scores": {
-        "bfcl.overall.rank": 85.0,
-        "bfcl.overall.overall_accuracy": 25.83,
-        "bfcl.overall.total_cost_usd": 7.49,
-        "bfcl.overall.latency_mean_s": 5.6,
-        "bfcl.overall.latency_std_s": 19.37,
-        "bfcl.overall.latency_p95_s": 22.6,
-        "bfcl.non_live.ast_accuracy": 84.0,
-        "bfcl.non_live.simple_ast_accuracy": 71.0,
-        "bfcl.non_live.multiple_ast_accuracy": 95.0,
-        "bfcl.non_live.parallel_ast_accuracy": 87.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 82.5,
-        "bfcl.live.live_accuracy": 70.76,
-        "bfcl.live.live_simple_ast_accuracy": 72.87,
-        "bfcl.live.live_multiple_ast_accuracy": 71.13,
-        "bfcl.live.live_parallel_ast_accuracy": 50.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 45.83,
-        "bfcl.multi_turn.accuracy": 11.12,
-        "bfcl.multi_turn.base_accuracy": 13.0,
-        "bfcl.multi_turn.miss_function_accuracy": 9.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 9.5,
-        "bfcl.multi_turn.long_context_accuracy": 13.0,
-        "bfcl.web_search.accuracy": 3.0,
-        "bfcl.web_search.base_accuracy": 6.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 10.75,
-        "bfcl.memory.kv_accuracy": 7.74,
-        "bfcl.memory.vector_accuracy": 5.81,
-        "bfcl.memory.recursive_summarization_accuracy": 18.71,
-        "bfcl.relevance.relevance_detection_accuracy": 93.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 42.7,
-        "bfcl.format_sensitivity.max_delta": 74.5,
-        "bfcl.format_sensitivity.stddev": 29.1
-      }
-    },
-    {
-      "model_id": "meta/llama-3-2-1b-instruct-fc",
-      "name": "Llama-3.2-1B-Instruct (FC)",
-      "developer": "Meta",
-      "scores": {
-        "bfcl.overall.rank": 107.0,
-        "bfcl.overall.overall_accuracy": 10.82,
-        "bfcl.overall.total_cost_usd": 1.64,
-        "bfcl.overall.latency_mean_s": 3.21,
-        "bfcl.overall.latency_std_s": 10.04,
-        "bfcl.overall.latency_p95_s": 9.77,
-        "bfcl.non_live.ast_accuracy": 38.38,
-        "bfcl.non_live.simple_ast_accuracy": 44.0,
-        "bfcl.non_live.multiple_ast_accuracy": 50.5,
-        "bfcl.non_live.parallel_ast_accuracy": 44.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 15.0,
-        "bfcl.live.live_accuracy": 11.77,
-        "bfcl.live.live_simple_ast_accuracy": 31.78,
-        "bfcl.live.live_multiple_ast_accuracy": 7.31,
-        "bfcl.live.live_parallel_ast_accuracy": 0.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 0.0,
-        "bfcl.multi_turn.accuracy": 0.0,
-        "bfcl.multi_turn.base_accuracy": 0.0,
-        "bfcl.multi_turn.miss_function_accuracy": 0.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 0.0,
-        "bfcl.multi_turn.long_context_accuracy": 0.0,
-        "bfcl.web_search.accuracy": 0.0,
-        "bfcl.web_search.base_accuracy": 0.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 3.23,
-        "bfcl.memory.kv_accuracy": 2.58,
-        "bfcl.memory.vector_accuracy": 2.58,
-        "bfcl.memory.recursive_summarization_accuracy": 4.52,
-        "bfcl.relevance.relevance_detection_accuracy": 43.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 51.57
-      }
-    },
-    {
-      "model_id": "meta/llama-3-2-3b-instruct-fc",
-      "name": "Llama-3.2-3B-Instruct (FC)",
-      "developer": "Meta",
-      "scores": {
-        "bfcl.overall.rank": 98.0,
-        "bfcl.overall.overall_accuracy": 21.95,
-        "bfcl.overall.total_cost_usd": 6.2,
-        "bfcl.overall.latency_mean_s": 6.1,
-        "bfcl.overall.latency_std_s": 20.07,
-        "bfcl.overall.latency_p95_s": 17.27,
-        "bfcl.non_live.ast_accuracy": 82.67,
-        "bfcl.non_live.simple_ast_accuracy": 70.67,
-        "bfcl.non_live.multiple_ast_accuracy": 92.5,
-        "bfcl.non_live.parallel_ast_accuracy": 88.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 79.0,
-        "bfcl.live.live_accuracy": 58.33,
-        "bfcl.live.live_simple_ast_accuracy": 65.12,
-        "bfcl.live.live_multiple_ast_accuracy": 57.64,
-        "bfcl.live.live_parallel_ast_accuracy": 25.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 37.5,
-        "bfcl.multi_turn.accuracy": 4.0,
-        "bfcl.multi_turn.base_accuracy": 5.0,
-        "bfcl.multi_turn.miss_function_accuracy": 3.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 4.0,
-        "bfcl.multi_turn.long_context_accuracy": 3.5,
-        "bfcl.web_search.accuracy": 1.0,
-        "bfcl.web_search.base_accuracy": 2.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 6.24,
-        "bfcl.memory.kv_accuracy": 3.23,
-        "bfcl.memory.vector_accuracy": 3.23,
-        "bfcl.memory.recursive_summarization_accuracy": 12.26,
-        "bfcl.relevance.relevance_detection_accuracy": 87.5,
-        "bfcl.relevance.irrelevance_detection_accuracy": 52.06
-      }
-    },
-    {
-      "model_id": "meta/llama-3-3-70b-instruct-fc",
-      "name": "Llama-3.3-70B-Instruct (FC)",
-      "developer": "Meta",
-      "scores": {
-        "bfcl.overall.rank": 62.0,
-        "bfcl.overall.overall_accuracy": 31.9,
-        "bfcl.overall.total_cost_usd": 29.54,
-        "bfcl.overall.latency_mean_s": 26.11,
-        "bfcl.overall.latency_std_s": 93.22,
-        "bfcl.overall.latency_p95_s": 187.93,
-        "bfcl.non_live.ast_accuracy": 88.02,
-        "bfcl.non_live.simple_ast_accuracy": 76.08,
-        "bfcl.non_live.multiple_ast_accuracy": 95.0,
-        "bfcl.non_live.parallel_ast_accuracy": 90.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 91.0,
-        "bfcl.live.live_accuracy": 76.61,
-        "bfcl.live.live_simple_ast_accuracy": 81.4,
-        "bfcl.live.live_multiple_ast_accuracy": 75.5,
-        "bfcl.live.live_parallel_ast_accuracy": 81.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 70.83,
-        "bfcl.multi_turn.accuracy": 21.5,
-        "bfcl.multi_turn.base_accuracy": 26.0,
-        "bfcl.multi_turn.miss_function_accuracy": 19.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 14.5,
-        "bfcl.multi_turn.long_context_accuracy": 26.5,
-        "bfcl.web_search.accuracy": 10.0,
-        "bfcl.web_search.base_accuracy": 14.0,
-        "bfcl.web_search.no_snippet_accuracy": 6.0,
-        "bfcl.memory.accuracy": 8.17,
-        "bfcl.memory.kv_accuracy": 4.52,
-        "bfcl.memory.vector_accuracy": 8.39,
-        "bfcl.memory.recursive_summarization_accuracy": 11.61,
-        "bfcl.relevance.relevance_detection_accuracy": 100.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 53.53
-      }
-    },
-    {
-      "model_id": "meta/llama-4-maverick-17b-128e-instruct-fp8-fc",
-      "name": "Llama-4-Maverick-17B-128E-Instruct-FP8 (FC)",
-      "developer": "Meta",
-      "scores": {
-        "bfcl.overall.rank": 50.0,
-        "bfcl.overall.overall_accuracy": 37.29,
-        "bfcl.overall.total_cost_usd": 18.25,
-        "bfcl.overall.latency_mean_s": 18.43,
-        "bfcl.overall.latency_std_s": 34.11,
-        "bfcl.overall.latency_p95_s": 102.75,
-        "bfcl.non_live.ast_accuracy": 88.65,
-        "bfcl.non_live.simple_ast_accuracy": 77.08,
-        "bfcl.non_live.multiple_ast_accuracy": 95.0,
-        "bfcl.non_live.parallel_ast_accuracy": 94.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 88.5,
-        "bfcl.live.live_accuracy": 73.65,
-        "bfcl.live.live_simple_ast_accuracy": 84.5,
-        "bfcl.live.live_multiple_ast_accuracy": 71.04,
-        "bfcl.live.live_parallel_ast_accuracy": 75.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 70.83,
-        "bfcl.multi_turn.accuracy": 20.25,
-        "bfcl.multi_turn.base_accuracy": 27.0,
-        "bfcl.multi_turn.miss_function_accuracy": 22.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 14.0,
-        "bfcl.multi_turn.long_context_accuracy": 18.0,
-        "bfcl.web_search.accuracy": 28.0,
-        "bfcl.web_search.base_accuracy": 39.0,
-        "bfcl.web_search.no_snippet_accuracy": 17.0,
-        "bfcl.memory.accuracy": 18.92,
-        "bfcl.memory.kv_accuracy": 8.39,
-        "bfcl.memory.vector_accuracy": 32.9,
-        "bfcl.memory.recursive_summarization_accuracy": 15.48,
-        "bfcl.relevance.relevance_detection_accuracy": 100.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 55.97
-      }
-    },
-    {
-      "model_id": "meta/llama-4-scout-17b-16e-instruct-fc",
-      "name": "Llama-4-Scout-17B-16E-Instruct (FC)",
-      "developer": "Meta",
-      "scores": {
-        "bfcl.overall.rank": 72.0,
-        "bfcl.overall.overall_accuracy": 28.13,
-        "bfcl.overall.total_cost_usd": 24.68,
-        "bfcl.overall.latency_mean_s": 17.86,
-        "bfcl.overall.latency_std_s": 50.68,
-        "bfcl.overall.latency_p95_s": 166.2,
-        "bfcl.non_live.ast_accuracy": 89.38,
-        "bfcl.non_live.simple_ast_accuracy": 79.0,
-        "bfcl.non_live.multiple_ast_accuracy": 94.0,
-        "bfcl.non_live.parallel_ast_accuracy": 94.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 90.5,
-        "bfcl.live.live_accuracy": 74.69,
-        "bfcl.live.live_simple_ast_accuracy": 81.78,
-        "bfcl.live.live_multiple_ast_accuracy": 72.74,
-        "bfcl.live.live_parallel_ast_accuracy": 81.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 79.17,
-        "bfcl.multi_turn.accuracy": 9.0,
-        "bfcl.multi_turn.base_accuracy": 12.0,
-        "bfcl.multi_turn.miss_function_accuracy": 7.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 7.5,
-        "bfcl.multi_turn.long_context_accuracy": 9.5,
-        "bfcl.web_search.accuracy": 14.5,
-        "bfcl.web_search.base_accuracy": 18.0,
-        "bfcl.web_search.no_snippet_accuracy": 11.0,
-        "bfcl.memory.accuracy": 8.17,
-        "bfcl.memory.kv_accuracy": 2.58,
-        "bfcl.memory.vector_accuracy": 2.58,
-        "bfcl.memory.recursive_summarization_accuracy": 19.35,
-        "bfcl.relevance.relevance_detection_accuracy": 100.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 44.92
-      }
-    },
-    {
-      "model_id": "microsoft/phi-4-prompt",
-      "name": "Phi-4 (Prompt)",
-      "developer": "microsoft",
-      "scores": {
-        "bfcl.overall.rank": 70.0,
-        "bfcl.overall.overall_accuracy": 28.79,
-        "bfcl.overall.total_cost_usd": 8.72,
-        "bfcl.overall.latency_mean_s": 9.49,
-        "bfcl.overall.latency_std_s": 26.73,
-        "bfcl.overall.latency_p95_s": 23.02,
-        "bfcl.non_live.ast_accuracy": 69.56,
-        "bfcl.non_live.simple_ast_accuracy": 74.25,
-        "bfcl.non_live.multiple_ast_accuracy": 89.5,
-        "bfcl.non_live.parallel_ast_accuracy": 65.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 49.5,
-        "bfcl.live.live_accuracy": 60.7,
-        "bfcl.live.live_simple_ast_accuracy": 65.5,
-        "bfcl.live.live_multiple_ast_accuracy": 59.64,
-        "bfcl.live.live_parallel_ast_accuracy": 81.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 41.67,
-        "bfcl.multi_turn.accuracy": 3.88,
-        "bfcl.multi_turn.base_accuracy": 9.0,
-        "bfcl.multi_turn.miss_function_accuracy": 0.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 3.5,
-        "bfcl.multi_turn.long_context_accuracy": 3.0,
-        "bfcl.web_search.accuracy": 4.5,
-        "bfcl.web_search.base_accuracy": 4.0,
-        "bfcl.web_search.no_snippet_accuracy": 5.0,
-        "bfcl.memory.accuracy": 24.73,
-        "bfcl.memory.kv_accuracy": 17.42,
-        "bfcl.memory.vector_accuracy": 25.16,
-        "bfcl.memory.recursive_summarization_accuracy": 31.61,
-        "bfcl.relevance.relevance_detection_accuracy": 50.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 87.55,
-        "bfcl.format_sensitivity.max_delta": 81.5,
-        "bfcl.format_sensitivity.stddev": 23.34
-      }
-    },
-    {
-      "model_id": "mistralai/ministral-8b-instruct-2410-fc",
-      "name": "Ministral-8B-Instruct-2410 (FC)",
-      "developer": "mistralai",
-      "scores": {
-        "bfcl.overall.rank": 105.0,
-        "bfcl.overall.overall_accuracy": 11.1,
-        "bfcl.overall.total_cost_usd": 70.01,
-        "bfcl.overall.latency_mean_s": 82.07,
-        "bfcl.overall.latency_std_s": 212.99,
-        "bfcl.overall.latency_p95_s": 568.59,
-        "bfcl.non_live.ast_accuracy": 0.0,
-        "bfcl.non_live.simple_ast_accuracy": 0.0,
-        "bfcl.non_live.multiple_ast_accuracy": 0.0,
-        "bfcl.non_live.parallel_ast_accuracy": 0.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 0.0,
-        "bfcl.live.live_accuracy": 0.0,
-        "bfcl.live.live_simple_ast_accuracy": 0.0,
-        "bfcl.live.live_multiple_ast_accuracy": 0.0,
-        "bfcl.live.live_parallel_ast_accuracy": 0.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 0.0,
-        "bfcl.multi_turn.accuracy": 0.0,
-        "bfcl.multi_turn.base_accuracy": 0.0,
-        "bfcl.multi_turn.miss_function_accuracy": 0.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 0.0,
-        "bfcl.multi_turn.long_context_accuracy": 0.0,
-        "bfcl.web_search.accuracy": 1.0,
-        "bfcl.web_search.base_accuracy": 2.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 4.52,
-        "bfcl.memory.kv_accuracy": 3.87,
-        "bfcl.memory.vector_accuracy": 7.1,
-        "bfcl.memory.recursive_summarization_accuracy": 2.58,
-        "bfcl.relevance.relevance_detection_accuracy": 0.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 100.0,
-        "bfcl.format_sensitivity.max_delta": 0.0,
-        "bfcl.format_sensitivity.stddev": 0.0
-      }
-    },
-    {
-      "model_id": "mistralai/mistral-large-2411-fc",
-      "name": "mistral-large-2411 (FC)",
-      "developer": "mistralai",
-      "scores": {
-        "bfcl.overall.rank": 46.0,
-        "bfcl.overall.overall_accuracy": 38.37,
-        "bfcl.overall.total_cost_usd": 115.98,
-        "bfcl.overall.latency_mean_s": 2.04,
-        "bfcl.overall.latency_std_s": 4.02,
-        "bfcl.overall.latency_p95_s": 4.68,
-        "bfcl.non_live.ast_accuracy": 84.65,
-        "bfcl.non_live.simple_ast_accuracy": 72.08,
-        "bfcl.non_live.multiple_ast_accuracy": 93.5,
-        "bfcl.non_live.parallel_ast_accuracy": 89.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 83.5,
-        "bfcl.live.live_accuracy": 81.87,
-        "bfcl.live.live_simple_ast_accuracy": 87.21,
-        "bfcl.live.live_multiple_ast_accuracy": 80.72,
-        "bfcl.live.live_parallel_ast_accuracy": 81.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 75.0,
-        "bfcl.multi_turn.accuracy": 14.12,
-        "bfcl.multi_turn.base_accuracy": 18.5,
-        "bfcl.multi_turn.miss_function_accuracy": 11.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 13.0,
-        "bfcl.multi_turn.long_context_accuracy": 13.5,
-        "bfcl.web_search.accuracy": 28.0,
-        "bfcl.web_search.base_accuracy": 41.0,
-        "bfcl.web_search.no_snippet_accuracy": 15.0,
-        "bfcl.memory.accuracy": 24.95,
-        "bfcl.memory.kv_accuracy": 18.71,
-        "bfcl.memory.vector_accuracy": 29.03,
-        "bfcl.memory.recursive_summarization_accuracy": 27.1,
-        "bfcl.relevance.relevance_detection_accuracy": 93.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 68.92
-      }
-    },
-    {
-      "model_id": "mistralai/mistral-large-2411-prompt",
-      "name": "mistral-large-2411 (Prompt)",
-      "developer": "mistralai",
-      "scores": {
-        "bfcl.overall.rank": 63.0,
-        "bfcl.overall.overall_accuracy": 31.84,
-        "bfcl.overall.total_cost_usd": 232.42,
-        "bfcl.overall.latency_mean_s": 1.82,
-        "bfcl.overall.latency_std_s": 7.15,
-        "bfcl.overall.latency_p95_s": 4.08,
-        "bfcl.non_live.ast_accuracy": 83.0,
-        "bfcl.non_live.simple_ast_accuracy": 75.5,
-        "bfcl.non_live.multiple_ast_accuracy": 89.5,
-        "bfcl.non_live.parallel_ast_accuracy": 87.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 80.0,
-        "bfcl.live.live_accuracy": 68.1,
-        "bfcl.live.live_simple_ast_accuracy": 83.72,
-        "bfcl.live.live_multiple_ast_accuracy": 64.01,
-        "bfcl.live.live_parallel_ast_accuracy": 93.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 62.5,
-        "bfcl.multi_turn.accuracy": 13.75,
-        "bfcl.multi_turn.base_accuracy": 20.0,
-        "bfcl.multi_turn.miss_function_accuracy": 5.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 11.0,
-        "bfcl.multi_turn.long_context_accuracy": 19.0,
-        "bfcl.web_search.accuracy": 20.0,
-        "bfcl.web_search.base_accuracy": 28.0,
-        "bfcl.web_search.no_snippet_accuracy": 12.0,
-        "bfcl.memory.accuracy": 23.66,
-        "bfcl.memory.kv_accuracy": 16.77,
-        "bfcl.memory.vector_accuracy": 30.97,
-        "bfcl.memory.recursive_summarization_accuracy": 23.23,
-        "bfcl.relevance.relevance_detection_accuracy": 93.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 38.77,
-        "bfcl.format_sensitivity.max_delta": 13.5,
-        "bfcl.format_sensitivity.stddev": 3.91
-      }
-    },
-    {
-      "model_id": "mistralai/mistral-medium-2505",
-      "name": "Mistral-Medium-2505",
-      "developer": "mistralai",
-      "scores": {
-        "bfcl.overall.rank": 48.0,
-        "bfcl.overall.overall_accuracy": 37.69,
-        "bfcl.overall.total_cost_usd": 36.51,
-        "bfcl.overall.latency_mean_s": 1.21,
-        "bfcl.overall.latency_std_s": 3.5,
-        "bfcl.overall.latency_p95_s": 2.86,
-        "bfcl.non_live.ast_accuracy": 85.33,
-        "bfcl.non_live.simple_ast_accuracy": 76.33,
-        "bfcl.non_live.multiple_ast_accuracy": 91.0,
-        "bfcl.non_live.parallel_ast_accuracy": 88.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 85.5,
-        "bfcl.live.live_accuracy": 66.03,
-        "bfcl.live.live_simple_ast_accuracy": 80.23,
-        "bfcl.live.live_multiple_ast_accuracy": 62.39,
-        "bfcl.live.live_parallel_ast_accuracy": 81.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 62.5,
-        "bfcl.multi_turn.accuracy": 9.88,
-        "bfcl.multi_turn.base_accuracy": 13.5,
-        "bfcl.multi_turn.miss_function_accuracy": 6.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 6.0,
-        "bfcl.multi_turn.long_context_accuracy": 13.5,
-        "bfcl.web_search.accuracy": 39.0,
-        "bfcl.web_search.base_accuracy": 41.0,
-        "bfcl.web_search.no_snippet_accuracy": 37.0,
-        "bfcl.memory.accuracy": 21.72,
-        "bfcl.memory.kv_accuracy": 16.13,
-        "bfcl.memory.vector_accuracy": 14.84,
-        "bfcl.memory.recursive_summarization_accuracy": 34.19,
-        "bfcl.relevance.relevance_detection_accuracy": 75.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 74.49,
-        "bfcl.format_sensitivity.max_delta": 21.5,
-        "bfcl.format_sensitivity.stddev": 5.02
-      }
-    },
-    {
-      "model_id": "mistralai/mistral-medium-2505-fc",
-      "name": "Mistral-Medium-2505 (FC)",
-      "developer": "mistralai",
-      "scores": {
-        "bfcl.overall.rank": 49.0,
-        "bfcl.overall.overall_accuracy": 37.56,
-        "bfcl.overall.total_cost_usd": 18.8,
-        "bfcl.overall.latency_mean_s": 1.6,
-        "bfcl.overall.latency_std_s": 4.44,
-        "bfcl.overall.latency_p95_s": 4.19,
-        "bfcl.non_live.ast_accuracy": 67.44,
-        "bfcl.non_live.simple_ast_accuracy": 39.75,
-        "bfcl.non_live.multiple_ast_accuracy": 78.0,
-        "bfcl.non_live.parallel_ast_accuracy": 83.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 69.0,
-        "bfcl.live.live_accuracy": 67.95,
-        "bfcl.live.live_simple_ast_accuracy": 67.05,
-        "bfcl.live.live_multiple_ast_accuracy": 68.09,
-        "bfcl.live.live_parallel_ast_accuracy": 81.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 62.5,
-        "bfcl.multi_turn.accuracy": 10.75,
-        "bfcl.multi_turn.base_accuracy": 15.5,
-        "bfcl.multi_turn.miss_function_accuracy": 7.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 7.5,
-        "bfcl.multi_turn.long_context_accuracy": 13.0,
-        "bfcl.web_search.accuracy": 35.0,
-        "bfcl.web_search.base_accuracy": 36.0,
-        "bfcl.web_search.no_snippet_accuracy": 34.0,
-        "bfcl.memory.accuracy": 23.01,
-        "bfcl.memory.kv_accuracy": 15.48,
-        "bfcl.memory.vector_accuracy": 20.0,
-        "bfcl.memory.recursive_summarization_accuracy": 33.55,
-        "bfcl.relevance.relevance_detection_accuracy": 62.5,
-        "bfcl.relevance.irrelevance_detection_accuracy": 91.95
-      }
-    },
-    {
-      "model_id": "mistralai/mistral-small-2506-fc",
-      "name": "Mistral-small-2506 (FC)",
-      "developer": "mistralai",
-      "scores": {
-        "bfcl.overall.rank": 51.0,
-        "bfcl.overall.overall_accuracy": 37.15,
-        "bfcl.overall.total_cost_usd": 5.2,
-        "bfcl.overall.latency_mean_s": 1.48,
-        "bfcl.overall.latency_std_s": 18.25,
-        "bfcl.overall.latency_p95_s": 2.5,
-        "bfcl.non_live.ast_accuracy": 73.6,
-        "bfcl.non_live.simple_ast_accuracy": 38.92,
-        "bfcl.non_live.multiple_ast_accuracy": 93.5,
-        "bfcl.non_live.parallel_ast_accuracy": 83.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 78.5,
-        "bfcl.live.live_accuracy": 77.28,
-        "bfcl.live.live_simple_ast_accuracy": 69.38,
-        "bfcl.live.live_multiple_ast_accuracy": 79.39,
-        "bfcl.live.live_parallel_ast_accuracy": 75.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 70.83,
-        "bfcl.multi_turn.accuracy": 11.5,
-        "bfcl.multi_turn.base_accuracy": 17.5,
-        "bfcl.multi_turn.miss_function_accuracy": 6.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 10.5,
-        "bfcl.multi_turn.long_context_accuracy": 12.0,
-        "bfcl.web_search.accuracy": 31.0,
-        "bfcl.web_search.base_accuracy": 37.0,
-        "bfcl.web_search.no_snippet_accuracy": 25.0,
-        "bfcl.memory.accuracy": 18.06,
-        "bfcl.memory.kv_accuracy": 8.39,
-        "bfcl.memory.vector_accuracy": 14.19,
-        "bfcl.memory.recursive_summarization_accuracy": 31.61,
-        "bfcl.relevance.relevance_detection_accuracy": 87.5,
-        "bfcl.relevance.irrelevance_detection_accuracy": 87.94
-      }
-    },
-    {
-      "model_id": "mistralai/mistral-small-2506-prompt",
-      "name": "Mistral-Small-2506 (Prompt)",
-      "developer": "mistralai",
-      "scores": {
-        "bfcl.overall.rank": 59.0,
-        "bfcl.overall.overall_accuracy": 32.38,
-        "bfcl.overall.total_cost_usd": 6.91,
-        "bfcl.overall.latency_mean_s": 0.92,
-        "bfcl.overall.latency_std_s": 6.79,
-        "bfcl.overall.latency_p95_s": 2.02,
-        "bfcl.non_live.ast_accuracy": 89.69,
-        "bfcl.non_live.simple_ast_accuracy": 78.75,
-        "bfcl.non_live.multiple_ast_accuracy": 96.0,
-        "bfcl.non_live.parallel_ast_accuracy": 92.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 91.5,
-        "bfcl.live.live_accuracy": 79.05,
-        "bfcl.live.live_simple_ast_accuracy": 81.4,
-        "bfcl.live.live_multiple_ast_accuracy": 78.54,
-        "bfcl.live.live_parallel_ast_accuracy": 93.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 66.67,
-        "bfcl.multi_turn.accuracy": 14.75,
-        "bfcl.multi_turn.base_accuracy": 20.5,
-        "bfcl.multi_turn.miss_function_accuracy": 17.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 9.5,
-        "bfcl.multi_turn.long_context_accuracy": 12.0,
-        "bfcl.web_search.accuracy": 7.5,
-        "bfcl.web_search.base_accuracy": 9.0,
-        "bfcl.web_search.no_snippet_accuracy": 6.0,
-        "bfcl.memory.accuracy": 15.05,
-        "bfcl.memory.kv_accuracy": 2.58,
-        "bfcl.memory.vector_accuracy": 11.61,
-        "bfcl.memory.recursive_summarization_accuracy": 30.97,
-        "bfcl.relevance.relevance_detection_accuracy": 93.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 65.73,
-        "bfcl.format_sensitivity.max_delta": 50.0,
-        "bfcl.format_sensitivity.stddev": 13.57
-      }
-    },
-    {
-      "model_id": "mistralai/open-mistral-nemo-2407-fc",
-      "name": "Open-Mistral-Nemo-2407 (FC)",
-      "developer": "mistralai",
-      "scores": {
-        "bfcl.overall.rank": 78.0,
-        "bfcl.overall.overall_accuracy": 27.63,
-        "bfcl.overall.total_cost_usd": 8.12,
-        "bfcl.overall.latency_mean_s": 1.07,
-        "bfcl.overall.latency_std_s": 11.93,
-        "bfcl.overall.latency_p95_s": 1.39,
-        "bfcl.non_live.ast_accuracy": 82.81,
-        "bfcl.non_live.simple_ast_accuracy": 65.75,
-        "bfcl.non_live.multiple_ast_accuracy": 92.5,
-        "bfcl.non_live.parallel_ast_accuracy": 88.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 84.5,
-        "bfcl.live.live_accuracy": 73.8,
-        "bfcl.live.live_simple_ast_accuracy": 78.68,
-        "bfcl.live.live_multiple_ast_accuracy": 72.84,
-        "bfcl.live.live_parallel_ast_accuracy": 68.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 66.67,
-        "bfcl.multi_turn.accuracy": 7.75,
-        "bfcl.multi_turn.base_accuracy": 12.5,
-        "bfcl.multi_turn.miss_function_accuracy": 6.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 7.5,
-        "bfcl.multi_turn.long_context_accuracy": 4.5,
-        "bfcl.web_search.accuracy": 7.0,
-        "bfcl.web_search.base_accuracy": 9.0,
-        "bfcl.web_search.no_snippet_accuracy": 5.0,
-        "bfcl.memory.accuracy": 10.32,
-        "bfcl.memory.kv_accuracy": 8.39,
-        "bfcl.memory.vector_accuracy": 9.68,
-        "bfcl.memory.recursive_summarization_accuracy": 12.9,
-        "bfcl.relevance.relevance_detection_accuracy": 81.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 61.77
-      }
-    },
-    {
-      "model_id": "mistralai/open-mistral-nemo-2407-prompt",
-      "name": "Open-Mistral-Nemo-2407 (Prompt)",
-      "developer": "mistralai",
-      "scores": {
-        "bfcl.overall.rank": 102.0,
-        "bfcl.overall.overall_accuracy": 19.31,
-        "bfcl.overall.total_cost_usd": 13.8,
-        "bfcl.overall.latency_mean_s": 0.84,
-        "bfcl.overall.latency_std_s": 7.05,
-        "bfcl.overall.latency_p95_s": 1.32,
-        "bfcl.non_live.ast_accuracy": 88.46,
-        "bfcl.non_live.simple_ast_accuracy": 79.33,
-        "bfcl.non_live.multiple_ast_accuracy": 92.5,
-        "bfcl.non_live.parallel_ast_accuracy": 90.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 91.5,
-        "bfcl.live.live_accuracy": 73.95,
-        "bfcl.live.live_simple_ast_accuracy": 78.29,
-        "bfcl.live.live_multiple_ast_accuracy": 73.03,
-        "bfcl.live.live_parallel_ast_accuracy": 87.5,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 58.33,
-        "bfcl.multi_turn.accuracy": 0.75,
-        "bfcl.multi_turn.base_accuracy": 0.5,
-        "bfcl.multi_turn.miss_function_accuracy": 1.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 0.0,
-        "bfcl.multi_turn.long_context_accuracy": 1.5,
-        "bfcl.web_search.accuracy": 2.5,
-        "bfcl.web_search.base_accuracy": 3.0,
-        "bfcl.web_search.no_snippet_accuracy": 2.0,
-        "bfcl.memory.accuracy": 8.6,
-        "bfcl.memory.kv_accuracy": 9.68,
-        "bfcl.memory.vector_accuracy": 9.68,
-        "bfcl.memory.recursive_summarization_accuracy": 6.45,
-        "bfcl.relevance.relevance_detection_accuracy": 93.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 6.28,
-        "bfcl.format_sensitivity.max_delta": 14.5,
-        "bfcl.format_sensitivity.stddev": 4.6
-      }
-    },
-    {
-      "model_id": "moonshotai/moonshotai-kimi-k2-instruct-fc",
-      "name": "Moonshotai-Kimi-K2-Instruct (FC)",
-      "developer": "moonshotai",
-      "scores": {
-        "bfcl.overall.rank": 11.0,
-        "bfcl.overall.overall_accuracy": 59.06,
-        "bfcl.overall.total_cost_usd": 6.19,
-        "bfcl.overall.latency_mean_s": 6.4,
-        "bfcl.overall.latency_std_s": 9.38,
-        "bfcl.overall.latency_p95_s": 13.78,
-        "bfcl.non_live.ast_accuracy": 81.6,
-        "bfcl.non_live.simple_ast_accuracy": 69.42,
-        "bfcl.non_live.multiple_ast_accuracy": 92.0,
-        "bfcl.non_live.parallel_ast_accuracy": 82.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 83.0,
-        "bfcl.live.live_accuracy": 78.68,
-        "bfcl.live.live_simple_ast_accuracy": 81.78,
-        "bfcl.live.live_multiple_ast_accuracy": 78.06,
-        "bfcl.live.live_parallel_ast_accuracy": 87.5,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 66.67,
-        "bfcl.multi_turn.accuracy": 50.63,
-        "bfcl.multi_turn.base_accuracy": 62.0,
-        "bfcl.multi_turn.miss_function_accuracy": 41.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 44.5,
-        "bfcl.multi_turn.long_context_accuracy": 55.0,
-        "bfcl.web_search.accuracy": 66.5,
-        "bfcl.web_search.base_accuracy": 72.0,
-        "bfcl.web_search.no_snippet_accuracy": 61.0,
-        "bfcl.memory.accuracy": 29.03,
-        "bfcl.memory.kv_accuracy": 21.94,
-        "bfcl.memory.vector_accuracy": 20.0,
-        "bfcl.memory.recursive_summarization_accuracy": 45.16,
-        "bfcl.relevance.relevance_detection_accuracy": 75.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 87.34
-      }
-    },
-    {
-      "model_id": "nanbeige/nanbeige3-5-pro-thinking-fc",
-      "name": "Nanbeige3.5-Pro-Thinking (FC)",
-      "developer": "nanbeige",
-      "scores": {
-        "bfcl.overall.rank": 32.0,
-        "bfcl.overall.overall_accuracy": 47.68,
-        "bfcl.overall.total_cost_usd": 23.46,
-        "bfcl.overall.latency_mean_s": 21.12,
-        "bfcl.overall.latency_std_s": 28.61,
-        "bfcl.overall.latency_p95_s": 63.29,
-        "bfcl.non_live.ast_accuracy": 38.35,
-        "bfcl.non_live.simple_ast_accuracy": 43.92,
-        "bfcl.non_live.multiple_ast_accuracy": 36.5,
-        "bfcl.non_live.parallel_ast_accuracy": 53.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 20.0,
-        "bfcl.live.live_accuracy": 69.95,
-        "bfcl.live.live_simple_ast_accuracy": 63.18,
-        "bfcl.live.live_multiple_ast_accuracy": 71.42,
-        "bfcl.live.live_parallel_ast_accuracy": 87.5,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 66.67,
-        "bfcl.multi_turn.accuracy": 40.0,
-        "bfcl.multi_turn.base_accuracy": 56.0,
-        "bfcl.multi_turn.miss_function_accuracy": 34.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 29.0,
-        "bfcl.multi_turn.long_context_accuracy": 41.0,
-        "bfcl.web_search.accuracy": 42.0,
-        "bfcl.web_search.base_accuracy": 47.0,
-        "bfcl.web_search.no_snippet_accuracy": 37.0,
-        "bfcl.memory.accuracy": 45.16,
-        "bfcl.memory.kv_accuracy": 38.06,
-        "bfcl.memory.vector_accuracy": 58.06,
-        "bfcl.memory.recursive_summarization_accuracy": 39.35,
-        "bfcl.relevance.relevance_detection_accuracy": 100.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 74.2
-      }
-    },
-    {
-      "model_id": "nanbeige/nanbeige4-3b-thinking-2511-fc",
-      "name": "Nanbeige4-3B-Thinking-2511 (FC)",
-      "developer": "nanbeige",
-      "scores": {
-        "bfcl.overall.rank": 25.0,
-        "bfcl.overall.overall_accuracy": 51.4,
-        "bfcl.overall.total_cost_usd": 14.14,
-        "bfcl.overall.latency_mean_s": 13.46,
-        "bfcl.overall.latency_std_s": 26.41,
-        "bfcl.overall.latency_p95_s": 37.45,
-        "bfcl.non_live.ast_accuracy": 81.58,
-        "bfcl.non_live.simple_ast_accuracy": 63.83,
-        "bfcl.non_live.multiple_ast_accuracy": 93.5,
-        "bfcl.non_live.parallel_ast_accuracy": 84.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 84.5,
-        "bfcl.live.live_accuracy": 79.42,
-        "bfcl.live.live_simple_ast_accuracy": 86.05,
-        "bfcl.live.live_multiple_ast_accuracy": 78.06,
-        "bfcl.live.live_parallel_ast_accuracy": 75.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 70.83,
-        "bfcl.multi_turn.accuracy": 51.12,
-        "bfcl.multi_turn.base_accuracy": 58.5,
-        "bfcl.multi_turn.miss_function_accuracy": 54.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 45.0,
-        "bfcl.multi_turn.long_context_accuracy": 47.0,
-        "bfcl.web_search.accuracy": 21.5,
-        "bfcl.web_search.base_accuracy": 31.0,
-        "bfcl.web_search.no_snippet_accuracy": 12.0,
-        "bfcl.memory.accuracy": 36.77,
-        "bfcl.memory.kv_accuracy": 31.61,
-        "bfcl.memory.vector_accuracy": 34.19,
-        "bfcl.memory.recursive_summarization_accuracy": 44.52,
-        "bfcl.relevance.relevance_detection_accuracy": 75.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 83.09
-      }
-    },
-    {
-      "model_id": "nvidia/llama-3-1-nemotron-ultra-253b-v1-fc",
-      "name": "Llama-3.1-Nemotron-Ultra-253B-v1 (FC)",
-      "developer": "nvidia",
-      "scores": {
-        "bfcl.overall.rank": 108.0,
-        "bfcl.overall.overall_accuracy": 10.0,
-        "bfcl.overall.total_cost_usd": 0.72,
-        "bfcl.overall.latency_mean_s": 1.42,
-        "bfcl.overall.latency_std_s": 1.84,
-        "bfcl.overall.latency_p95_s": 2.4,
-        "bfcl.non_live.ast_accuracy": 0.0,
-        "bfcl.non_live.simple_ast_accuracy": 0.0,
-        "bfcl.non_live.multiple_ast_accuracy": 0.0,
-        "bfcl.non_live.parallel_ast_accuracy": 0.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 0.0,
-        "bfcl.live.live_accuracy": 0.0,
-        "bfcl.live.live_simple_ast_accuracy": 0.0,
-        "bfcl.live.live_multiple_ast_accuracy": 0.0,
-        "bfcl.live.live_parallel_ast_accuracy": 0.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 0.0,
-        "bfcl.multi_turn.accuracy": 0.0,
-        "bfcl.multi_turn.base_accuracy": 0.0,
-        "bfcl.multi_turn.miss_function_accuracy": 0.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 0.0,
-        "bfcl.multi_turn.long_context_accuracy": 0.0,
-        "bfcl.web_search.accuracy": 0.0,
-        "bfcl.web_search.base_accuracy": 0.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 0.0,
-        "bfcl.memory.kv_accuracy": 0.0,
-        "bfcl.memory.vector_accuracy": 0.0,
-        "bfcl.memory.recursive_summarization_accuracy": 0.0,
-        "bfcl.relevance.relevance_detection_accuracy": 0.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 100.0
-      }
-    },
-    {
-      "model_id": "openai/gpt-4-1-2025-04-14-fc",
-      "name": "GPT-4.1-2025-04-14 (FC)",
-      "developer": "OpenAI",
-      "scores": {
-        "bfcl.overall.rank": 20.0,
-        "bfcl.overall.overall_accuracy": 53.96,
-        "bfcl.overall.total_cost_usd": 100.75,
-        "bfcl.overall.latency_mean_s": 1.63,
-        "bfcl.overall.latency_std_s": 3.05,
-        "bfcl.overall.latency_p95_s": 4.01,
-        "bfcl.non_live.ast_accuracy": 82.79,
-        "bfcl.non_live.simple_ast_accuracy": 72.67,
-        "bfcl.non_live.multiple_ast_accuracy": 89.0,
-        "bfcl.non_live.parallel_ast_accuracy": 88.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 81.5,
-        "bfcl.live.live_accuracy": 69.95,
-        "bfcl.live.live_simple_ast_accuracy": 69.38,
-        "bfcl.live.live_multiple_ast_accuracy": 70.28,
-        "bfcl.live.live_parallel_ast_accuracy": 56.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 70.83,
-        "bfcl.multi_turn.accuracy": 38.88,
-        "bfcl.multi_turn.base_accuracy": 47.5,
-        "bfcl.multi_turn.miss_function_accuracy": 32.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 32.5,
-        "bfcl.multi_turn.long_context_accuracy": 43.0,
-        "bfcl.web_search.accuracy": 68.0,
-        "bfcl.web_search.base_accuracy": 67.0,
-        "bfcl.web_search.no_snippet_accuracy": 69.0,
-        "bfcl.memory.accuracy": 23.87,
-        "bfcl.memory.kv_accuracy": 16.13,
-        "bfcl.memory.vector_accuracy": 18.06,
-        "bfcl.memory.recursive_summarization_accuracy": 37.42,
-        "bfcl.relevance.relevance_detection_accuracy": 87.5,
-        "bfcl.relevance.irrelevance_detection_accuracy": 86.52
-      }
-    },
-    {
-      "model_id": "openai/gpt-4-1-2025-04-14-prompt",
-      "name": "GPT-4.1-2025-04-14 (Prompt)",
-      "developer": "OpenAI",
-      "scores": {
-        "bfcl.overall.rank": 45.0,
-        "bfcl.overall.overall_accuracy": 39.38,
-        "bfcl.overall.total_cost_usd": 145.85,
-        "bfcl.overall.latency_mean_s": 1.2,
-        "bfcl.overall.latency_std_s": 3.23,
-        "bfcl.overall.latency_p95_s": 2.53,
-        "bfcl.non_live.ast_accuracy": 88.69,
-        "bfcl.non_live.simple_ast_accuracy": 78.25,
-        "bfcl.non_live.multiple_ast_accuracy": 93.5,
-        "bfcl.non_live.parallel_ast_accuracy": 94.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 89.0,
-        "bfcl.live.live_accuracy": 78.9,
-        "bfcl.live.live_simple_ast_accuracy": 84.88,
-        "bfcl.live.live_multiple_ast_accuracy": 77.4,
-        "bfcl.live.live_parallel_ast_accuracy": 87.5,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 75.0,
-        "bfcl.multi_turn.accuracy": 9.75,
-        "bfcl.multi_turn.base_accuracy": 10.5,
-        "bfcl.multi_turn.miss_function_accuracy": 11.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 8.0,
-        "bfcl.multi_turn.long_context_accuracy": 9.5,
-        "bfcl.web_search.accuracy": 35.0,
-        "bfcl.web_search.base_accuracy": 40.0,
-        "bfcl.web_search.no_snippet_accuracy": 30.0,
-        "bfcl.memory.accuracy": 21.51,
-        "bfcl.memory.kv_accuracy": 9.68,
-        "bfcl.memory.vector_accuracy": 19.35,
-        "bfcl.memory.recursive_summarization_accuracy": 35.48,
-        "bfcl.relevance.relevance_detection_accuracy": 100.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 83.99,
-        "bfcl.format_sensitivity.max_delta": 23.5,
-        "bfcl.format_sensitivity.stddev": 6.18
-      }
-    },
-    {
-      "model_id": "openai/gpt-4-1-mini-2025-04-14-fc",
-      "name": "GPT-4.1-mini-2025-04-14 (FC)",
-      "developer": "OpenAI",
-      "scores": {
-        "bfcl.overall.rank": 27.0,
-        "bfcl.overall.overall_accuracy": 50.45,
-        "bfcl.overall.total_cost_usd": 19.25,
-        "bfcl.overall.latency_mean_s": 1.32,
-        "bfcl.overall.latency_std_s": 3.65,
-        "bfcl.overall.latency_p95_s": 2.4,
-        "bfcl.non_live.ast_accuracy": 83.83,
-        "bfcl.non_live.simple_ast_accuracy": 73.33,
-        "bfcl.non_live.multiple_ast_accuracy": 89.0,
-        "bfcl.non_live.parallel_ast_accuracy": 91.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 82.0,
-        "bfcl.live.live_accuracy": 68.84,
-        "bfcl.live.live_simple_ast_accuracy": 67.05,
-        "bfcl.live.live_multiple_ast_accuracy": 69.8,
-        "bfcl.live.live_parallel_ast_accuracy": 43.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 62.5,
-        "bfcl.multi_turn.accuracy": 34.13,
-        "bfcl.multi_turn.base_accuracy": 43.5,
-        "bfcl.multi_turn.miss_function_accuracy": 22.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 30.5,
-        "bfcl.multi_turn.long_context_accuracy": 40.0,
-        "bfcl.web_search.accuracy": 57.0,
-        "bfcl.web_search.base_accuracy": 62.0,
-        "bfcl.web_search.no_snippet_accuracy": 52.0,
-        "bfcl.memory.accuracy": 26.88,
-        "bfcl.memory.kv_accuracy": 22.58,
-        "bfcl.memory.vector_accuracy": 16.13,
-        "bfcl.memory.recursive_summarization_accuracy": 41.94,
-        "bfcl.relevance.relevance_detection_accuracy": 81.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 81.69
-      }
-    },
-    {
-      "model_id": "openai/gpt-4-1-mini-2025-04-14-prompt",
-      "name": "GPT-4.1-mini-2025-04-14 (Prompt)",
-      "developer": "OpenAI",
-      "scores": {
-        "bfcl.overall.rank": 67.0,
-        "bfcl.overall.overall_accuracy": 29.73,
-        "bfcl.overall.total_cost_usd": 20.52,
-        "bfcl.overall.latency_mean_s": 1.36,
-        "bfcl.overall.latency_std_s": 4.5,
-        "bfcl.overall.latency_p95_s": 3.38,
-        "bfcl.non_live.ast_accuracy": 84.6,
-        "bfcl.non_live.simple_ast_accuracy": 74.92,
-        "bfcl.non_live.multiple_ast_accuracy": 92.5,
-        "bfcl.non_live.parallel_ast_accuracy": 87.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 83.5,
-        "bfcl.live.live_accuracy": 74.76,
-        "bfcl.live.live_simple_ast_accuracy": 80.62,
-        "bfcl.live.live_multiple_ast_accuracy": 73.31,
-        "bfcl.live.live_parallel_ast_accuracy": 81.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 70.83,
-        "bfcl.multi_turn.accuracy": 2.5,
-        "bfcl.multi_turn.base_accuracy": 1.5,
-        "bfcl.multi_turn.miss_function_accuracy": 4.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 2.5,
-        "bfcl.multi_turn.long_context_accuracy": 1.5,
-        "bfcl.web_search.accuracy": 4.0,
-        "bfcl.web_search.base_accuracy": 7.0,
-        "bfcl.web_search.no_snippet_accuracy": 1.0,
-        "bfcl.memory.accuracy": 24.3,
-        "bfcl.memory.kv_accuracy": 20.65,
-        "bfcl.memory.vector_accuracy": 13.55,
-        "bfcl.memory.recursive_summarization_accuracy": 38.71,
-        "bfcl.relevance.relevance_detection_accuracy": 87.5,
-        "bfcl.relevance.irrelevance_detection_accuracy": 73.88,
-        "bfcl.format_sensitivity.max_delta": 45.0,
-        "bfcl.format_sensitivity.stddev": 13.33
-      }
-    },
-    {
-      "model_id": "openai/gpt-4-1-nano-2025-04-14-fc",
-      "name": "GPT-4.1-nano-2025-04-14 (FC)",
-      "developer": "OpenAI",
-      "scores": {
-        "bfcl.overall.rank": 58.0,
-        "bfcl.overall.overall_accuracy": 33.05,
-        "bfcl.overall.total_cost_usd": 5.66,
-        "bfcl.overall.latency_mean_s": 1.44,
-        "bfcl.overall.latency_std_s": 10.84,
-        "bfcl.overall.latency_p95_s": 2.26,
-        "bfcl.non_live.ast_accuracy": 72.98,
-        "bfcl.non_live.simple_ast_accuracy": 59.92,
-        "bfcl.non_live.multiple_ast_accuracy": 79.5,
-        "bfcl.non_live.parallel_ast_accuracy": 84.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 68.5,
-        "bfcl.live.live_accuracy": 60.77,
-        "bfcl.live.live_simple_ast_accuracy": 58.14,
-        "bfcl.live.live_multiple_ast_accuracy": 61.44,
-        "bfcl.live.live_parallel_ast_accuracy": 68.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 54.17,
-        "bfcl.multi_turn.accuracy": 23.62,
-        "bfcl.multi_turn.base_accuracy": 39.5,
-        "bfcl.multi_turn.miss_function_accuracy": 7.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 17.5,
-        "bfcl.multi_turn.long_context_accuracy": 30.0,
-        "bfcl.web_search.accuracy": 11.0,
-        "bfcl.web_search.base_accuracy": 13.0,
-        "bfcl.web_search.no_snippet_accuracy": 9.0,
-        "bfcl.memory.accuracy": 18.92,
-        "bfcl.memory.kv_accuracy": 10.32,
-        "bfcl.memory.vector_accuracy": 19.35,
-        "bfcl.memory.recursive_summarization_accuracy": 27.1,
-        "bfcl.relevance.relevance_detection_accuracy": 93.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 66.0
-      }
-    },
-    {
-      "model_id": "openai/gpt-4-1-nano-2025-04-14-prompt",
-      "name": "GPT-4.1-nano-2025-04-14 (Prompt)",
-      "developer": "OpenAI",
-      "scores": {
-        "bfcl.overall.rank": 90.0,
-        "bfcl.overall.overall_accuracy": 24.88,
-        "bfcl.overall.total_cost_usd": 7.42,
-        "bfcl.overall.latency_mean_s": 1.02,
-        "bfcl.overall.latency_std_s": 7.3,
-        "bfcl.overall.latency_p95_s": 1.88,
-        "bfcl.non_live.ast_accuracy": 72.44,
-        "bfcl.non_live.simple_ast_accuracy": 68.75,
-        "bfcl.non_live.multiple_ast_accuracy": 63.5,
-        "bfcl.non_live.parallel_ast_accuracy": 85.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 72.5,
-        "bfcl.live.live_accuracy": 50.33,
-        "bfcl.live.live_simple_ast_accuracy": 63.18,
-        "bfcl.live.live_multiple_ast_accuracy": 46.53,
-        "bfcl.live.live_parallel_ast_accuracy": 87.5,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 54.17,
-        "bfcl.multi_turn.accuracy": 2.0,
-        "bfcl.multi_turn.base_accuracy": 2.5,
-        "bfcl.multi_turn.miss_function_accuracy": 1.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 2.5,
-        "bfcl.multi_turn.long_context_accuracy": 2.0,
-        "bfcl.web_search.accuracy": 1.5,
-        "bfcl.web_search.base_accuracy": 2.0,
-        "bfcl.web_search.no_snippet_accuracy": 1.0,
-        "bfcl.memory.accuracy": 16.77,
-        "bfcl.memory.kv_accuracy": 9.03,
-        "bfcl.memory.vector_accuracy": 14.19,
-        "bfcl.memory.recursive_summarization_accuracy": 27.1,
-        "bfcl.relevance.relevance_detection_accuracy": 68.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 83.44,
-        "bfcl.format_sensitivity.max_delta": 73.0,
-        "bfcl.format_sensitivity.stddev": 17.08
-      }
-    },
-    {
-      "model_id": "openai/gpt-5-2-2025-12-11-fc",
-      "name": "GPT-5.2-2025-12-11 (FC)",
-      "developer": "OpenAI",
-      "scores": {
-        "bfcl.overall.rank": 16.0,
-        "bfcl.overall.overall_accuracy": 55.87,
-        "bfcl.overall.total_cost_usd": 85.65,
-        "bfcl.overall.latency_mean_s": 2.23,
-        "bfcl.overall.latency_std_s": 9.75,
-        "bfcl.overall.latency_p95_s": 5.26,
-        "bfcl.non_live.ast_accuracy": 81.85,
-        "bfcl.non_live.simple_ast_accuracy": 72.92,
-        "bfcl.non_live.multiple_ast_accuracy": 88.0,
-        "bfcl.non_live.parallel_ast_accuracy": 89.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 77.5,
-        "bfcl.live.live_accuracy": 70.39,
-        "bfcl.live.live_simple_ast_accuracy": 71.71,
-        "bfcl.live.live_multiple_ast_accuracy": 70.37,
-        "bfcl.live.live_parallel_ast_accuracy": 68.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 58.33,
-        "bfcl.multi_turn.accuracy": 28.12,
-        "bfcl.multi_turn.base_accuracy": 36.5,
-        "bfcl.multi_turn.miss_function_accuracy": 18.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 27.5,
-        "bfcl.multi_turn.long_context_accuracy": 30.5,
-        "bfcl.web_search.accuracy": 75.5,
-        "bfcl.web_search.base_accuracy": 78.0,
-        "bfcl.web_search.no_snippet_accuracy": 73.0,
-        "bfcl.memory.accuracy": 45.81,
-        "bfcl.memory.kv_accuracy": 33.55,
-        "bfcl.memory.vector_accuracy": 43.23,
-        "bfcl.memory.recursive_summarization_accuracy": 60.65,
-        "bfcl.relevance.relevance_detection_accuracy": 75.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 79.42
-      }
-    },
-    {
-      "model_id": "openai/gpt-5-2-2025-12-11-prompt",
-      "name": "GPT-5.2-2025-12-11 (Prompt)",
-      "developer": "OpenAI",
-      "scores": {
-        "bfcl.overall.rank": 38.0,
-        "bfcl.overall.overall_accuracy": 45.27,
-        "bfcl.overall.total_cost_usd": 164.58,
-        "bfcl.overall.latency_mean_s": 4.21,
-        "bfcl.overall.latency_std_s": 20.93,
-        "bfcl.overall.latency_p95_s": 10.58,
-        "bfcl.non_live.ast_accuracy": 78.29,
-        "bfcl.non_live.simple_ast_accuracy": 71.17,
-        "bfcl.non_live.multiple_ast_accuracy": 83.5,
-        "bfcl.non_live.parallel_ast_accuracy": 84.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 74.5,
-        "bfcl.live.live_accuracy": 67.14,
-        "bfcl.live.live_simple_ast_accuracy": 77.91,
-        "bfcl.live.live_multiple_ast_accuracy": 64.58,
-        "bfcl.live.live_parallel_ast_accuracy": 75.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 58.33,
-        "bfcl.multi_turn.accuracy": 43.75,
-        "bfcl.multi_turn.base_accuracy": 54.5,
-        "bfcl.multi_turn.miss_function_accuracy": 40.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 33.5,
-        "bfcl.multi_turn.long_context_accuracy": 46.5,
-        "bfcl.web_search.accuracy": 40.5,
-        "bfcl.web_search.base_accuracy": 45.0,
-        "bfcl.web_search.no_snippet_accuracy": 36.0,
-        "bfcl.memory.accuracy": 3.87,
-        "bfcl.memory.kv_accuracy": 2.58,
-        "bfcl.memory.vector_accuracy": 1.94,
-        "bfcl.memory.recursive_summarization_accuracy": 7.1,
-        "bfcl.relevance.relevance_detection_accuracy": 75.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 87.26,
-        "bfcl.format_sensitivity.max_delta": 13.0,
-        "bfcl.format_sensitivity.stddev": 3.25
-      }
-    },
-    {
-      "model_id": "openai/gpt-5-mini-2025-08-07-fc",
-      "name": "GPT-5-mini-2025-08-07 (FC)",
-      "developer": "OpenAI",
-      "scores": {
-        "bfcl.overall.rank": 17.0,
-        "bfcl.overall.overall_accuracy": 55.46,
-        "bfcl.overall.total_cost_usd": 22.18,
-        "bfcl.overall.latency_mean_s": 8.32,
-        "bfcl.overall.latency_std_s": 17.35,
-        "bfcl.overall.latency_p95_s": 19.8,
-        "bfcl.non_live.ast_accuracy": 69.85,
-        "bfcl.non_live.simple_ast_accuracy": 59.92,
-        "bfcl.non_live.multiple_ast_accuracy": 69.0,
-        "bfcl.non_live.parallel_ast_accuracy": 80.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 70.5,
-        "bfcl.live.live_accuracy": 58.62,
-        "bfcl.live.live_simple_ast_accuracy": 62.02,
-        "bfcl.live.live_multiple_ast_accuracy": 58.02,
-        "bfcl.live.live_parallel_ast_accuracy": 62.5,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 45.83,
-        "bfcl.multi_turn.accuracy": 27.5,
-        "bfcl.multi_turn.base_accuracy": 36.5,
-        "bfcl.multi_turn.miss_function_accuracy": 17.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 23.5,
-        "bfcl.multi_turn.long_context_accuracy": 33.0,
-        "bfcl.web_search.accuracy": 82.0,
-        "bfcl.web_search.base_accuracy": 87.0,
-        "bfcl.web_search.no_snippet_accuracy": 77.0,
-        "bfcl.memory.accuracy": 44.3,
-        "bfcl.memory.kv_accuracy": 36.77,
-        "bfcl.memory.vector_accuracy": 43.87,
-        "bfcl.memory.recursive_summarization_accuracy": 52.26,
-        "bfcl.relevance.relevance_detection_accuracy": 62.5,
-        "bfcl.relevance.irrelevance_detection_accuracy": 91.01
-      }
-    },
-    {
-      "model_id": "openai/gpt-5-mini-2025-08-07-prompt",
-      "name": "GPT-5-mini-2025-08-07 (Prompt)",
-      "developer": "OpenAI",
-      "scores": {
-        "bfcl.overall.rank": 77.0,
-        "bfcl.overall.overall_accuracy": 27.83,
-        "bfcl.overall.total_cost_usd": 82.74,
-        "bfcl.overall.latency_mean_s": 8.89,
-        "bfcl.overall.latency_std_s": 11.08,
-        "bfcl.overall.latency_p95_s": 19.72,
-        "bfcl.non_live.ast_accuracy": 68.04,
-        "bfcl.non_live.simple_ast_accuracy": 59.17,
-        "bfcl.non_live.multiple_ast_accuracy": 72.5,
-        "bfcl.non_live.parallel_ast_accuracy": 71.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 69.0,
-        "bfcl.live.live_accuracy": 62.55,
-        "bfcl.live.live_simple_ast_accuracy": 69.77,
-        "bfcl.live.live_multiple_ast_accuracy": 61.16,
-        "bfcl.live.live_parallel_ast_accuracy": 75.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 37.5,
-        "bfcl.multi_turn.accuracy": 5.5,
-        "bfcl.multi_turn.base_accuracy": 5.5,
-        "bfcl.multi_turn.miss_function_accuracy": 5.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 4.5,
-        "bfcl.multi_turn.long_context_accuracy": 7.0,
-        "bfcl.web_search.accuracy": 8.5,
-        "bfcl.web_search.base_accuracy": 11.0,
-        "bfcl.web_search.no_snippet_accuracy": 6.0,
-        "bfcl.memory.accuracy": 29.25,
-        "bfcl.memory.kv_accuracy": 19.35,
-        "bfcl.memory.vector_accuracy": 29.68,
-        "bfcl.memory.recursive_summarization_accuracy": 38.71,
-        "bfcl.relevance.relevance_detection_accuracy": 93.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 55.71,
-        "bfcl.format_sensitivity.max_delta": 16.0,
-        "bfcl.format_sensitivity.stddev": 3.78
-      }
-    },
-    {
-      "model_id": "openai/gpt-5-nano-2025-08-07-fc",
-      "name": "GPT-5-nano-2025-08-07 (FC)",
-      "developer": "OpenAI",
-      "scores": {
-        "bfcl.overall.rank": 24.0,
-        "bfcl.overall.overall_accuracy": 51.45,
-        "bfcl.overall.total_cost_usd": 8.79,
-        "bfcl.overall.latency_mean_s": 10.36,
-        "bfcl.overall.latency_std_s": 10.37,
-        "bfcl.overall.latency_p95_s": 23.56,
-        "bfcl.non_live.ast_accuracy": 68.0,
-        "bfcl.non_live.simple_ast_accuracy": 57.0,
-        "bfcl.non_live.multiple_ast_accuracy": 64.5,
-        "bfcl.non_live.parallel_ast_accuracy": 79.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 71.5,
-        "bfcl.live.live_accuracy": 59.44,
-        "bfcl.live.live_simple_ast_accuracy": 58.91,
-        "bfcl.live.live_multiple_ast_accuracy": 59.83,
-        "bfcl.live.live_parallel_ast_accuracy": 50.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 54.17,
-        "bfcl.multi_turn.accuracy": 34.5,
-        "bfcl.multi_turn.base_accuracy": 44.0,
-        "bfcl.multi_turn.miss_function_accuracy": 23.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 32.5,
-        "bfcl.multi_turn.long_context_accuracy": 38.0,
-        "bfcl.web_search.accuracy": 72.5,
-        "bfcl.web_search.base_accuracy": 74.0,
-        "bfcl.web_search.no_snippet_accuracy": 71.0,
-        "bfcl.memory.accuracy": 24.73,
-        "bfcl.memory.kv_accuracy": 18.06,
-        "bfcl.memory.vector_accuracy": 27.1,
-        "bfcl.memory.recursive_summarization_accuracy": 29.03,
-        "bfcl.relevance.relevance_detection_accuracy": 75.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 89.1
-      }
-    },
-    {
-      "model_id": "openai/gpt-5-nano-2025-08-07-prompt",
-      "name": "GPT-5-nano-2025-08-07 (Prompt)",
-      "developer": "OpenAI",
-      "scores": {
-        "bfcl.overall.rank": 79.0,
-        "bfcl.overall.overall_accuracy": 27.55,
-        "bfcl.overall.total_cost_usd": 21.47,
-        "bfcl.overall.latency_mean_s": 10.67,
-        "bfcl.overall.latency_std_s": 7.68,
-        "bfcl.overall.latency_p95_s": 23.28,
-        "bfcl.non_live.ast_accuracy": 80.81,
-        "bfcl.non_live.simple_ast_accuracy": 69.25,
-        "bfcl.non_live.multiple_ast_accuracy": 86.0,
-        "bfcl.non_live.parallel_ast_accuracy": 87.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 80.5,
-        "bfcl.live.live_accuracy": 70.69,
-        "bfcl.live.live_simple_ast_accuracy": 76.36,
-        "bfcl.live.live_multiple_ast_accuracy": 69.71,
-        "bfcl.live.live_parallel_ast_accuracy": 68.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 54.17,
-        "bfcl.multi_turn.accuracy": 0.75,
-        "bfcl.multi_turn.base_accuracy": 1.0,
-        "bfcl.multi_turn.miss_function_accuracy": 1.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 0.0,
-        "bfcl.multi_turn.long_context_accuracy": 1.0,
-        "bfcl.web_search.accuracy": 13.5,
-        "bfcl.web_search.base_accuracy": 10.0,
-        "bfcl.web_search.no_snippet_accuracy": 17.0,
-        "bfcl.memory.accuracy": 24.52,
-        "bfcl.memory.kv_accuracy": 20.65,
-        "bfcl.memory.vector_accuracy": 31.61,
-        "bfcl.memory.recursive_summarization_accuracy": 21.29,
-        "bfcl.relevance.relevance_detection_accuracy": 93.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 45.75,
-        "bfcl.format_sensitivity.max_delta": 8.5,
-        "bfcl.format_sensitivity.stddev": 2.57
-      }
-    },
-    {
-      "model_id": "openai/o3-2025-04-16-fc",
-      "name": "o3-2025-04-16 (FC)",
-      "developer": "OpenAI",
-      "scores": {
-        "bfcl.overall.rank": 30.0,
-        "bfcl.overall.overall_accuracy": 48.56,
-        "bfcl.overall.total_cost_usd": 133.45,
-        "bfcl.overall.latency_mean_s": 3.5,
-        "bfcl.overall.latency_std_s": 8.69,
-        "bfcl.overall.latency_p95_s": 8.39,
-        "bfcl.non_live.ast_accuracy": 40.38,
-        "bfcl.non_live.simple_ast_accuracy": 74.5,
-        "bfcl.non_live.multiple_ast_accuracy": 87.0,
-        "bfcl.non_live.parallel_ast_accuracy": 0.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 0.0,
-        "bfcl.live.live_accuracy": 66.17,
-        "bfcl.live.live_simple_ast_accuracy": 70.54,
-        "bfcl.live.live_multiple_ast_accuracy": 67.62,
-        "bfcl.live.live_parallel_ast_accuracy": 0.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 0.0,
-        "bfcl.multi_turn.accuracy": 14.75,
-        "bfcl.multi_turn.base_accuracy": 16.5,
-        "bfcl.multi_turn.miss_function_accuracy": 11.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 14.5,
-        "bfcl.multi_turn.long_context_accuracy": 16.5,
-        "bfcl.web_search.accuracy": 77.0,
-        "bfcl.web_search.base_accuracy": 79.0,
-        "bfcl.web_search.no_snippet_accuracy": 75.0,
-        "bfcl.memory.accuracy": 47.31,
-        "bfcl.memory.kv_accuracy": 24.52,
-        "bfcl.memory.vector_accuracy": 44.52,
-        "bfcl.memory.recursive_summarization_accuracy": 72.9,
-        "bfcl.relevance.relevance_detection_accuracy": 81.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 86.13
-      }
-    },
-    {
-      "model_id": "openai/o3-2025-04-16-prompt",
-      "name": "o3-2025-04-16 (Prompt)",
-      "developer": "OpenAI",
-      "scores": {
-        "bfcl.overall.rank": 8.0,
-        "bfcl.overall.overall_accuracy": 63.05,
-        "bfcl.overall.total_cost_usd": 234.64,
-        "bfcl.overall.latency_mean_s": 4.83,
-        "bfcl.overall.latency_std_s": 7.01,
-        "bfcl.overall.latency_p95_s": 11.7,
-        "bfcl.non_live.ast_accuracy": 81.94,
-        "bfcl.non_live.simple_ast_accuracy": 74.25,
-        "bfcl.non_live.multiple_ast_accuracy": 89.0,
-        "bfcl.non_live.parallel_ast_accuracy": 86.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 78.0,
-        "bfcl.live.live_accuracy": 73.21,
-        "bfcl.live.live_simple_ast_accuracy": 83.33,
-        "bfcl.live.live_multiple_ast_accuracy": 70.75,
-        "bfcl.live.live_parallel_ast_accuracy": 75.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 70.83,
-        "bfcl.multi_turn.accuracy": 62.25,
-        "bfcl.multi_turn.base_accuracy": 68.0,
-        "bfcl.multi_turn.miss_function_accuracy": 63.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 54.5,
-        "bfcl.multi_turn.long_context_accuracy": 63.0,
-        "bfcl.web_search.accuracy": 50.5,
-        "bfcl.web_search.base_accuracy": 51.0,
-        "bfcl.web_search.no_snippet_accuracy": 50.0,
-        "bfcl.memory.accuracy": 51.83,
-        "bfcl.memory.kv_accuracy": 33.55,
-        "bfcl.memory.vector_accuracy": 50.32,
-        "bfcl.memory.recursive_summarization_accuracy": 71.61,
-        "bfcl.relevance.relevance_detection_accuracy": 93.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 83.98,
-        "bfcl.format_sensitivity.max_delta": 8.5,
-        "bfcl.format_sensitivity.stddev": 2.75
-      }
-    },
-    {
-      "model_id": "openai/o4-mini-2025-04-16-fc",
-      "name": "o4-mini-2025-04-16 (FC)",
-      "developer": "OpenAI",
-      "scores": {
-        "bfcl.overall.rank": 21.0,
-        "bfcl.overall.overall_accuracy": 53.24,
-        "bfcl.overall.total_cost_usd": 81.91,
-        "bfcl.overall.latency_mean_s": 3.71,
-        "bfcl.overall.latency_std_s": 7.18,
-        "bfcl.overall.latency_p95_s": 9.33,
-        "bfcl.non_live.ast_accuracy": 37.73,
-        "bfcl.non_live.simple_ast_accuracy": 66.92,
-        "bfcl.non_live.multiple_ast_accuracy": 84.0,
-        "bfcl.non_live.parallel_ast_accuracy": 0.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 0.0,
-        "bfcl.live.live_accuracy": 66.1,
-        "bfcl.live.live_simple_ast_accuracy": 69.38,
-        "bfcl.live.live_multiple_ast_accuracy": 67.81,
-        "bfcl.live.live_parallel_ast_accuracy": 0.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 0.0,
-        "bfcl.multi_turn.accuracy": 41.75,
-        "bfcl.multi_turn.base_accuracy": 51.0,
-        "bfcl.multi_turn.miss_function_accuracy": 30.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 40.5,
-        "bfcl.multi_turn.long_context_accuracy": 45.5,
-        "bfcl.web_search.accuracy": 75.5,
-        "bfcl.web_search.base_accuracy": 75.0,
-        "bfcl.web_search.no_snippet_accuracy": 76.0,
-        "bfcl.memory.accuracy": 34.19,
-        "bfcl.memory.kv_accuracy": 19.35,
-        "bfcl.memory.vector_accuracy": 24.52,
-        "bfcl.memory.recursive_summarization_accuracy": 58.71,
-        "bfcl.relevance.relevance_detection_accuracy": 81.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 83.91
-      }
-    },
-    {
-      "model_id": "openai/o4-mini-2025-04-16-prompt",
-      "name": "o4-mini-2025-04-16 (Prompt)",
-      "developer": "OpenAI",
-      "scores": {
-        "bfcl.overall.rank": 28.0,
-        "bfcl.overall.overall_accuracy": 50.26,
-        "bfcl.overall.total_cost_usd": 133.63,
-        "bfcl.overall.latency_mean_s": 4.47,
-        "bfcl.overall.latency_std_s": 5.19,
-        "bfcl.overall.latency_p95_s": 10.19,
-        "bfcl.non_live.ast_accuracy": 81.29,
-        "bfcl.non_live.simple_ast_accuracy": 72.67,
-        "bfcl.non_live.multiple_ast_accuracy": 88.0,
-        "bfcl.non_live.parallel_ast_accuracy": 84.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 80.0,
-        "bfcl.live.live_accuracy": 70.76,
-        "bfcl.live.live_simple_ast_accuracy": 79.46,
-        "bfcl.live.live_multiple_ast_accuracy": 68.76,
-        "bfcl.live.live_parallel_ast_accuracy": 75.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 62.5,
-        "bfcl.multi_turn.accuracy": 16.62,
-        "bfcl.multi_turn.base_accuracy": 16.5,
-        "bfcl.multi_turn.miss_function_accuracy": 18.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 17.5,
-        "bfcl.multi_turn.long_context_accuracy": 14.5,
-        "bfcl.web_search.accuracy": 71.5,
-        "bfcl.web_search.base_accuracy": 73.0,
-        "bfcl.web_search.no_snippet_accuracy": 70.0,
-        "bfcl.memory.accuracy": 35.27,
-        "bfcl.memory.kv_accuracy": 22.58,
-        "bfcl.memory.vector_accuracy": 25.16,
-        "bfcl.memory.recursive_summarization_accuracy": 58.06,
-        "bfcl.relevance.relevance_detection_accuracy": 81.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 87.16,
-        "bfcl.format_sensitivity.max_delta": 9.5,
-        "bfcl.format_sensitivity.stddev": 2.6
-      }
-    },
-    {
-      "model_id": "openbmb/minicpm3-4b-fc-fc",
-      "name": "MiniCPM3-4B-FC (FC)",
-      "developer": "openbmb",
-      "scores": {
-        "bfcl.overall.rank": 86.0,
-        "bfcl.overall.overall_accuracy": 25.55,
-        "bfcl.overall.total_cost_usd": 54.05,
-        "bfcl.overall.latency_mean_s": 118.62,
-        "bfcl.overall.latency_std_s": 143.98,
-        "bfcl.overall.latency_p95_s": 388.67,
-        "bfcl.non_live.ast_accuracy": 81.75,
-        "bfcl.non_live.simple_ast_accuracy": 70.5,
-        "bfcl.non_live.multiple_ast_accuracy": 92.0,
-        "bfcl.non_live.parallel_ast_accuracy": 84.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 80.5,
-        "bfcl.live.live_accuracy": 65.21,
-        "bfcl.live.live_simple_ast_accuracy": 73.26,
-        "bfcl.live.live_multiple_ast_accuracy": 63.53,
-        "bfcl.live.live_parallel_ast_accuracy": 50.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 62.5,
-        "bfcl.multi_turn.accuracy": 3.88,
-        "bfcl.multi_turn.base_accuracy": 6.5,
-        "bfcl.multi_turn.miss_function_accuracy": 2.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 4.5,
-        "bfcl.multi_turn.long_context_accuracy": 2.5,
-        "bfcl.web_search.accuracy": 0.0,
-        "bfcl.web_search.base_accuracy": 0.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 12.04,
-        "bfcl.memory.kv_accuracy": 9.68,
-        "bfcl.memory.vector_accuracy": 15.48,
-        "bfcl.memory.recursive_summarization_accuracy": 10.97,
-        "bfcl.relevance.relevance_detection_accuracy": 68.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 72.84
-      }
-    },
-    {
-      "model_id": "openbmb/minicpm3-4b-prompt",
-      "name": "MiniCPM3-4B (Prompt)",
-      "developer": "openbmb",
-      "scores": {
-        "bfcl.overall.rank": 97.0,
-        "bfcl.overall.overall_accuracy": 22.08,
-        "bfcl.overall.total_cost_usd": 29.83,
-        "bfcl.overall.latency_mean_s": 31.18,
-        "bfcl.overall.latency_std_s": 35.61,
-        "bfcl.overall.latency_p95_s": 102.02,
-        "bfcl.non_live.ast_accuracy": 70.54,
-        "bfcl.non_live.simple_ast_accuracy": 66.17,
-        "bfcl.non_live.multiple_ast_accuracy": 77.0,
-        "bfcl.non_live.parallel_ast_accuracy": 70.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 69.0,
-        "bfcl.live.live_accuracy": 43.15,
-        "bfcl.live.live_simple_ast_accuracy": 47.67,
-        "bfcl.live.live_multiple_ast_accuracy": 42.17,
-        "bfcl.live.live_parallel_ast_accuracy": 43.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 37.5,
-        "bfcl.multi_turn.accuracy": 3.5,
-        "bfcl.multi_turn.base_accuracy": 4.5,
-        "bfcl.multi_turn.miss_function_accuracy": 4.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 2.0,
-        "bfcl.multi_turn.long_context_accuracy": 3.0,
-        "bfcl.web_search.accuracy": 2.0,
-        "bfcl.web_search.base_accuracy": 2.0,
-        "bfcl.web_search.no_snippet_accuracy": 2.0,
-        "bfcl.memory.accuracy": 9.46,
-        "bfcl.memory.kv_accuracy": 8.39,
-        "bfcl.memory.vector_accuracy": 10.32,
-        "bfcl.memory.recursive_summarization_accuracy": 9.68,
-        "bfcl.relevance.relevance_detection_accuracy": 56.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 73.71,
-        "bfcl.format_sensitivity.max_delta": 68.0,
-        "bfcl.format_sensitivity.stddev": 16.55
-      }
-    },
-    {
-      "model_id": "phronetic-ai/rzn-t-prompt",
-      "name": "RZN-T (Prompt)",
-      "developer": "phronetic-ai",
-      "scores": {
-        "bfcl.overall.rank": 96.0,
-        "bfcl.overall.overall_accuracy": 22.25,
-        "bfcl.overall.total_cost_usd": 12.31,
-        "bfcl.overall.latency_mean_s": 12.32,
-        "bfcl.overall.latency_std_s": 27.53,
-        "bfcl.overall.latency_p95_s": 39.84,
-        "bfcl.non_live.ast_accuracy": 67.94,
-        "bfcl.non_live.simple_ast_accuracy": 63.75,
-        "bfcl.non_live.multiple_ast_accuracy": 75.0,
-        "bfcl.non_live.parallel_ast_accuracy": 69.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 63.5,
-        "bfcl.live.live_accuracy": 49.74,
-        "bfcl.live.live_simple_ast_accuracy": 61.24,
-        "bfcl.live.live_multiple_ast_accuracy": 47.2,
-        "bfcl.live.live_parallel_ast_accuracy": 43.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 41.67,
-        "bfcl.multi_turn.accuracy": 2.88,
-        "bfcl.multi_turn.base_accuracy": 4.5,
-        "bfcl.multi_turn.miss_function_accuracy": 2.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 2.5,
-        "bfcl.multi_turn.long_context_accuracy": 2.5,
-        "bfcl.web_search.accuracy": 0.0,
-        "bfcl.web_search.base_accuracy": 0.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 6.88,
-        "bfcl.memory.kv_accuracy": 9.03,
-        "bfcl.memory.vector_accuracy": 6.45,
-        "bfcl.memory.recursive_summarization_accuracy": 5.16,
-        "bfcl.relevance.relevance_detection_accuracy": 81.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 82.41,
-        "bfcl.format_sensitivity.max_delta": 63.5,
-        "bfcl.format_sensitivity.stddev": 25.53
-      }
-    },
-    {
-      "model_id": "qwen/qwen3-0-6b-fc",
-      "name": "Qwen3-0.6B (FC)",
-      "developer": "qwen",
-      "scores": {
-        "bfcl.overall.rank": 92.0,
-        "bfcl.overall.overall_accuracy": 23.93,
-        "bfcl.overall.total_cost_usd": 0.46,
-        "bfcl.overall.latency_mean_s": 0.68,
-        "bfcl.overall.latency_std_s": 8.45,
-        "bfcl.overall.latency_p95_s": 0.96,
-        "bfcl.non_live.ast_accuracy": 71.79,
-        "bfcl.non_live.simple_ast_accuracy": 64.17,
-        "bfcl.non_live.multiple_ast_accuracy": 86.0,
-        "bfcl.non_live.parallel_ast_accuracy": 67.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 69.5,
-        "bfcl.live.live_accuracy": 56.62,
-        "bfcl.live.live_simple_ast_accuracy": 61.24,
-        "bfcl.live.live_multiple_ast_accuracy": 56.13,
-        "bfcl.live.live_parallel_ast_accuracy": 43.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 37.5,
-        "bfcl.multi_turn.accuracy": 3.62,
-        "bfcl.multi_turn.base_accuracy": 5.5,
-        "bfcl.multi_turn.miss_function_accuracy": 2.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 3.0,
-        "bfcl.multi_turn.long_context_accuracy": 4.0,
-        "bfcl.web_search.accuracy": 1.0,
-        "bfcl.web_search.base_accuracy": 1.0,
-        "bfcl.web_search.no_snippet_accuracy": 1.0,
-        "bfcl.memory.accuracy": 8.6,
-        "bfcl.memory.kv_accuracy": 2.58,
-        "bfcl.memory.vector_accuracy": 1.94,
-        "bfcl.memory.recursive_summarization_accuracy": 21.29,
-        "bfcl.relevance.relevance_detection_accuracy": 75.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 80.84
-      }
-    },
-    {
-      "model_id": "qwen/qwen3-0-6b-prompt",
-      "name": "Qwen3-0.6B (Prompt)",
-      "developer": "qwen",
-      "scores": {
-        "bfcl.overall.rank": 94.0,
-        "bfcl.overall.overall_accuracy": 22.38,
-        "bfcl.overall.total_cost_usd": 3.65,
-        "bfcl.overall.latency_mean_s": 3.1,
-        "bfcl.overall.latency_std_s": 4.32,
-        "bfcl.overall.latency_p95_s": 10.31,
-        "bfcl.non_live.ast_accuracy": 70.0,
-        "bfcl.non_live.simple_ast_accuracy": 64.0,
-        "bfcl.non_live.multiple_ast_accuracy": 78.0,
-        "bfcl.non_live.parallel_ast_accuracy": 75.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 63.0,
-        "bfcl.live.live_accuracy": 49.37,
-        "bfcl.live.live_simple_ast_accuracy": 57.75,
-        "bfcl.live.live_multiple_ast_accuracy": 47.77,
-        "bfcl.live.live_parallel_ast_accuracy": 37.5,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 37.5,
-        "bfcl.multi_turn.accuracy": 1.38,
-        "bfcl.multi_turn.base_accuracy": 1.5,
-        "bfcl.multi_turn.miss_function_accuracy": 1.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 1.5,
-        "bfcl.multi_turn.long_context_accuracy": 1.0,
-        "bfcl.web_search.accuracy": 0.5,
-        "bfcl.web_search.base_accuracy": 1.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 8.39,
-        "bfcl.memory.kv_accuracy": 1.29,
-        "bfcl.memory.vector_accuracy": 2.58,
-        "bfcl.memory.recursive_summarization_accuracy": 21.29,
-        "bfcl.relevance.relevance_detection_accuracy": 75.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 82.5,
-        "bfcl.format_sensitivity.max_delta": 60.5,
-        "bfcl.format_sensitivity.stddev": 24.35
-      }
-    },
-    {
-      "model_id": "qwen/qwen3-1-7b-fc",
-      "name": "Qwen3-1.7B (FC)",
-      "developer": "qwen",
-      "scores": {
-        "bfcl.overall.rank": 71.0,
-        "bfcl.overall.overall_accuracy": 28.41,
-        "bfcl.overall.total_cost_usd": 4.33,
-        "bfcl.overall.latency_mean_s": 5.12,
-        "bfcl.overall.latency_std_s": 7.37,
-        "bfcl.overall.latency_p95_s": 13.35,
-        "bfcl.non_live.ast_accuracy": 82.92,
-        "bfcl.non_live.simple_ast_accuracy": 70.67,
-        "bfcl.non_live.multiple_ast_accuracy": 92.5,
-        "bfcl.non_live.parallel_ast_accuracy": 88.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 80.0,
-        "bfcl.live.live_accuracy": 74.61,
-        "bfcl.live.live_simple_ast_accuracy": 76.74,
-        "bfcl.live.live_multiple_ast_accuracy": 74.26,
-        "bfcl.live.live_parallel_ast_accuracy": 62.5,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 75.0,
-        "bfcl.multi_turn.accuracy": 11.0,
-        "bfcl.multi_turn.base_accuracy": 15.0,
-        "bfcl.multi_turn.miss_function_accuracy": 6.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 12.0,
-        "bfcl.multi_turn.long_context_accuracy": 11.0,
-        "bfcl.web_search.accuracy": 2.5,
-        "bfcl.web_search.base_accuracy": 3.0,
-        "bfcl.web_search.no_snippet_accuracy": 2.0,
-        "bfcl.memory.accuracy": 6.02,
-        "bfcl.memory.kv_accuracy": 4.52,
-        "bfcl.memory.vector_accuracy": 7.74,
-        "bfcl.memory.recursive_summarization_accuracy": 5.81,
-        "bfcl.relevance.relevance_detection_accuracy": 81.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 76.54
-      }
-    },
-    {
-      "model_id": "qwen/qwen3-14b-fc",
-      "name": "Qwen3-14B (FC)",
-      "developer": "qwen",
-      "scores": {
-        "bfcl.overall.rank": 43.0,
-        "bfcl.overall.overall_accuracy": 41.03,
-        "bfcl.overall.total_cost_usd": 3.38,
-        "bfcl.overall.latency_mean_s": 4.5,
-        "bfcl.overall.latency_std_s": 18.84,
-        "bfcl.overall.latency_p95_s": 13.34,
-        "bfcl.non_live.ast_accuracy": 84.94,
-        "bfcl.non_live.simple_ast_accuracy": 74.75,
-        "bfcl.non_live.multiple_ast_accuracy": 93.0,
-        "bfcl.non_live.parallel_ast_accuracy": 80.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 92.0,
-        "bfcl.live.live_accuracy": 80.01,
-        "bfcl.live.live_simple_ast_accuracy": 85.66,
-        "bfcl.live.live_multiple_ast_accuracy": 79.01,
-        "bfcl.live.live_parallel_ast_accuracy": 68.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 70.83,
-        "bfcl.multi_turn.accuracy": 34.75,
-        "bfcl.multi_turn.base_accuracy": 39.0,
-        "bfcl.multi_turn.miss_function_accuracy": 34.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 33.5,
-        "bfcl.multi_turn.long_context_accuracy": 32.5,
-        "bfcl.web_search.accuracy": 10.0,
-        "bfcl.web_search.base_accuracy": 8.0,
-        "bfcl.web_search.no_snippet_accuracy": 12.0,
-        "bfcl.memory.accuracy": 19.57,
-        "bfcl.memory.kv_accuracy": 7.1,
-        "bfcl.memory.vector_accuracy": 16.77,
-        "bfcl.memory.recursive_summarization_accuracy": 34.84,
-        "bfcl.relevance.relevance_detection_accuracy": 87.5,
-        "bfcl.relevance.irrelevance_detection_accuracy": 81.94
-      }
-    },
-    {
-      "model_id": "qwen/qwen3-14b-prompt",
-      "name": "Qwen3-14B (Prompt)",
-      "developer": "qwen",
-      "scores": {
-        "bfcl.overall.rank": 47.0,
-        "bfcl.overall.overall_accuracy": 37.77,
-        "bfcl.overall.total_cost_usd": 1.35,
-        "bfcl.overall.latency_mean_s": 1.2,
-        "bfcl.overall.latency_std_s": 8.5,
-        "bfcl.overall.latency_p95_s": 2.3,
-        "bfcl.non_live.ast_accuracy": 89.46,
-        "bfcl.non_live.simple_ast_accuracy": 76.83,
-        "bfcl.non_live.multiple_ast_accuracy": 93.5,
-        "bfcl.non_live.parallel_ast_accuracy": 95.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 92.0,
-        "bfcl.live.live_accuracy": 79.35,
-        "bfcl.live.live_simple_ast_accuracy": 84.5,
-        "bfcl.live.live_multiple_ast_accuracy": 78.06,
-        "bfcl.live.live_parallel_ast_accuracy": 87.5,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 75.0,
-        "bfcl.multi_turn.accuracy": 26.13,
-        "bfcl.multi_turn.base_accuracy": 16.5,
-        "bfcl.multi_turn.miss_function_accuracy": 37.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 31.0,
-        "bfcl.multi_turn.long_context_accuracy": 19.5,
-        "bfcl.web_search.accuracy": 10.5,
-        "bfcl.web_search.base_accuracy": 6.0,
-        "bfcl.web_search.no_snippet_accuracy": 15.0,
-        "bfcl.memory.accuracy": 11.18,
-        "bfcl.memory.kv_accuracy": 4.52,
-        "bfcl.memory.vector_accuracy": 6.45,
-        "bfcl.memory.recursive_summarization_accuracy": 22.58,
-        "bfcl.relevance.relevance_detection_accuracy": 81.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 87.18,
-        "bfcl.format_sensitivity.max_delta": 14.0,
-        "bfcl.format_sensitivity.stddev": 3.97
-      }
-    },
-    {
-      "model_id": "qwen/qwen3-235b-a22b-instruct-2507-fc",
-      "name": "Qwen3-235B-A22B-Instruct-2507 (FC)",
-      "developer": "qwen",
-      "scores": {
-        "bfcl.overall.rank": 31.0,
-        "bfcl.overall.overall_accuracy": 47.99,
-        "bfcl.overall.total_cost_usd": 2.5,
-        "bfcl.overall.latency_mean_s": 2.57,
-        "bfcl.overall.latency_std_s": 2.44,
-        "bfcl.overall.latency_p95_s": 6.27,
-        "bfcl.non_live.ast_accuracy": 37.4,
-        "bfcl.non_live.simple_ast_accuracy": 40.58,
-        "bfcl.non_live.multiple_ast_accuracy": 36.5,
-        "bfcl.non_live.parallel_ast_accuracy": 53.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 19.5,
-        "bfcl.live.live_accuracy": 68.91,
-        "bfcl.live.live_simple_ast_accuracy": 58.53,
-        "bfcl.live.live_multiple_ast_accuracy": 71.6,
-        "bfcl.live.live_parallel_ast_accuracy": 68.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 62.5,
-        "bfcl.multi_turn.accuracy": 45.38,
-        "bfcl.multi_turn.base_accuracy": 57.5,
-        "bfcl.multi_turn.miss_function_accuracy": 35.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 33.5,
-        "bfcl.multi_turn.long_context_accuracy": 55.5,
-        "bfcl.web_search.accuracy": 54.0,
-        "bfcl.web_search.base_accuracy": 57.0,
-        "bfcl.web_search.no_snippet_accuracy": 51.0,
-        "bfcl.memory.accuracy": 23.87,
-        "bfcl.memory.kv_accuracy": 7.1,
-        "bfcl.memory.vector_accuracy": 18.71,
-        "bfcl.memory.recursive_summarization_accuracy": 45.81,
-        "bfcl.relevance.relevance_detection_accuracy": 87.5,
-        "bfcl.relevance.irrelevance_detection_accuracy": 81.73
-      }
-    },
-    {
-      "model_id": "qwen/qwen3-235b-a22b-instruct-2507-prompt",
-      "name": "Qwen3-235B-A22B-Instruct-2507 (Prompt)",
-      "developer": "qwen",
-      "scores": {
-        "bfcl.overall.rank": 23.0,
-        "bfcl.overall.overall_accuracy": 52.15,
-        "bfcl.overall.total_cost_usd": 3.12,
-        "bfcl.overall.latency_mean_s": 2.56,
-        "bfcl.overall.latency_std_s": 2.75,
-        "bfcl.overall.latency_p95_s": 7.61,
-        "bfcl.non_live.ast_accuracy": 90.33,
-        "bfcl.non_live.simple_ast_accuracy": 79.83,
-        "bfcl.non_live.multiple_ast_accuracy": 95.0,
-        "bfcl.non_live.parallel_ast_accuracy": 95.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 91.0,
-        "bfcl.live.live_accuracy": 78.68,
-        "bfcl.live.live_simple_ast_accuracy": 82.95,
-        "bfcl.live.live_multiple_ast_accuracy": 77.78,
-        "bfcl.live.live_parallel_ast_accuracy": 81.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 70.83,
-        "bfcl.multi_turn.accuracy": 44.62,
-        "bfcl.multi_turn.base_accuracy": 54.0,
-        "bfcl.multi_turn.miss_function_accuracy": 42.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 31.5,
-        "bfcl.multi_turn.long_context_accuracy": 50.5,
-        "bfcl.web_search.accuracy": 50.5,
-        "bfcl.web_search.base_accuracy": 56.0,
-        "bfcl.web_search.no_snippet_accuracy": 45.0,
-        "bfcl.memory.accuracy": 19.35,
-        "bfcl.memory.kv_accuracy": 12.9,
-        "bfcl.memory.vector_accuracy": 11.61,
-        "bfcl.memory.recursive_summarization_accuracy": 33.55,
-        "bfcl.relevance.relevance_detection_accuracy": 93.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 78.89,
-        "bfcl.format_sensitivity.max_delta": 8.0,
-        "bfcl.format_sensitivity.stddev": 1.95
-      }
-    },
-    {
-      "model_id": "qwen/qwen3-30b-a3b-instruct-2507-fc",
-      "name": "Qwen3-30B-A3B-Instruct-2507 (FC)",
-      "developer": "qwen",
-      "scores": {
-        "bfcl.overall.rank": 41.0,
-        "bfcl.overall.overall_accuracy": 41.39,
-        "bfcl.overall.total_cost_usd": 5.62,
-        "bfcl.overall.latency_mean_s": 5.95,
-        "bfcl.overall.latency_std_s": 25.48,
-        "bfcl.overall.latency_p95_s": 12.7,
-        "bfcl.non_live.ast_accuracy": 85.77,
-        "bfcl.non_live.simple_ast_accuracy": 68.58,
-        "bfcl.non_live.multiple_ast_accuracy": 94.5,
-        "bfcl.non_live.parallel_ast_accuracy": 91.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 88.5,
-        "bfcl.live.live_accuracy": 77.94,
-        "bfcl.live.live_simple_ast_accuracy": 83.33,
-        "bfcl.live.live_multiple_ast_accuracy": 76.83,
-        "bfcl.live.live_parallel_ast_accuracy": 68.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 75.0,
-        "bfcl.multi_turn.accuracy": 30.0,
-        "bfcl.multi_turn.base_accuracy": 43.5,
-        "bfcl.multi_turn.miss_function_accuracy": 10.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 25.0,
-        "bfcl.multi_turn.long_context_accuracy": 41.0,
-        "bfcl.web_search.accuracy": 22.5,
-        "bfcl.web_search.base_accuracy": 21.0,
-        "bfcl.web_search.no_snippet_accuracy": 24.0,
-        "bfcl.memory.accuracy": 17.63,
-        "bfcl.memory.kv_accuracy": 9.03,
-        "bfcl.memory.vector_accuracy": 9.03,
-        "bfcl.memory.recursive_summarization_accuracy": 34.84,
-        "bfcl.relevance.relevance_detection_accuracy": 81.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 79.9
-      }
-    },
-    {
-      "model_id": "qwen/qwen3-30b-a3b-instruct-2507-prompt",
-      "name": "Qwen3-30B-A3B-Instruct-2507 (Prompt)",
-      "developer": "qwen",
-      "scores": {
-        "bfcl.overall.rank": 53.0,
-        "bfcl.overall.overall_accuracy": 36.7,
-        "bfcl.overall.total_cost_usd": 1.56,
-        "bfcl.overall.latency_mean_s": 1.24,
-        "bfcl.overall.latency_std_s": 7.9,
-        "bfcl.overall.latency_p95_s": 2.84,
-        "bfcl.non_live.ast_accuracy": 88.92,
-        "bfcl.non_live.simple_ast_accuracy": 80.67,
-        "bfcl.non_live.multiple_ast_accuracy": 93.0,
-        "bfcl.non_live.parallel_ast_accuracy": 94.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 88.0,
-        "bfcl.live.live_accuracy": 78.39,
-        "bfcl.live.live_simple_ast_accuracy": 82.56,
-        "bfcl.live.live_multiple_ast_accuracy": 77.49,
-        "bfcl.live.live_parallel_ast_accuracy": 87.5,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 66.67,
-        "bfcl.multi_turn.accuracy": 23.5,
-        "bfcl.multi_turn.base_accuracy": 33.0,
-        "bfcl.multi_turn.miss_function_accuracy": 16.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 16.0,
-        "bfcl.multi_turn.long_context_accuracy": 29.0,
-        "bfcl.web_search.accuracy": 17.5,
-        "bfcl.web_search.base_accuracy": 15.0,
-        "bfcl.web_search.no_snippet_accuracy": 20.0,
-        "bfcl.memory.accuracy": 9.68,
-        "bfcl.memory.kv_accuracy": 5.81,
-        "bfcl.memory.vector_accuracy": 6.45,
-        "bfcl.memory.recursive_summarization_accuracy": 16.77,
-        "bfcl.relevance.relevance_detection_accuracy": 93.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 74.85,
-        "bfcl.format_sensitivity.max_delta": 16.0,
-        "bfcl.format_sensitivity.stddev": 4.13
-      }
-    },
-    {
-      "model_id": "qwen/qwen3-32b-fc",
-      "name": "Qwen3-32B (FC)",
-      "developer": "qwen",
-      "scores": {
-        "bfcl.overall.rank": 29.0,
-        "bfcl.overall.overall_accuracy": 48.71,
-        "bfcl.overall.total_cost_usd": 153.08,
-        "bfcl.overall.latency_mean_s": 169.87,
-        "bfcl.overall.latency_std_s": 164.27,
-        "bfcl.overall.latency_p95_s": 473.49,
-        "bfcl.non_live.ast_accuracy": 88.77,
-        "bfcl.non_live.simple_ast_accuracy": 75.58,
-        "bfcl.non_live.multiple_ast_accuracy": 94.5,
-        "bfcl.non_live.parallel_ast_accuracy": 93.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 91.5,
-        "bfcl.live.live_accuracy": 82.01,
-        "bfcl.live.live_simple_ast_accuracy": 89.53,
-        "bfcl.live.live_multiple_ast_accuracy": 80.91,
-        "bfcl.live.live_parallel_ast_accuracy": 81.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 50.0,
-        "bfcl.multi_turn.accuracy": 47.87,
-        "bfcl.multi_turn.base_accuracy": 56.0,
-        "bfcl.multi_turn.miss_function_accuracy": 52.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 40.0,
-        "bfcl.multi_turn.long_context_accuracy": 43.0,
-        "bfcl.web_search.accuracy": 21.5,
-        "bfcl.web_search.base_accuracy": 25.0,
-        "bfcl.web_search.no_snippet_accuracy": 18.0,
-        "bfcl.memory.accuracy": 26.67,
-        "bfcl.memory.kv_accuracy": 12.26,
-        "bfcl.memory.vector_accuracy": 25.81,
-        "bfcl.memory.recursive_summarization_accuracy": 41.94,
-        "bfcl.relevance.relevance_detection_accuracy": 93.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 76.37
-      }
-    },
-    {
-      "model_id": "qwen/qwen3-32b-prompt",
-      "name": "Qwen3-32B (Prompt)",
-      "developer": "qwen",
-      "scores": {
-        "bfcl.overall.rank": 33.0,
-        "bfcl.overall.overall_accuracy": 46.78,
-        "bfcl.overall.total_cost_usd": 199.47,
-        "bfcl.overall.latency_mean_s": 167.54,
-        "bfcl.overall.latency_std_s": 160.5,
-        "bfcl.overall.latency_p95_s": 457.87,
-        "bfcl.non_live.ast_accuracy": 90.27,
-        "bfcl.non_live.simple_ast_accuracy": 79.08,
-        "bfcl.non_live.multiple_ast_accuracy": 97.0,
-        "bfcl.non_live.parallel_ast_accuracy": 93.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 91.5,
-        "bfcl.live.live_accuracy": 82.01,
-        "bfcl.live.live_simple_ast_accuracy": 87.21,
-        "bfcl.live.live_multiple_ast_accuracy": 81.2,
-        "bfcl.live.live_parallel_ast_accuracy": 81.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 62.5,
-        "bfcl.multi_turn.accuracy": 43.25,
-        "bfcl.multi_turn.base_accuracy": 54.0,
-        "bfcl.multi_turn.miss_function_accuracy": 46.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 36.5,
-        "bfcl.multi_turn.long_context_accuracy": 36.5,
-        "bfcl.web_search.accuracy": 26.0,
-        "bfcl.web_search.base_accuracy": 34.0,
-        "bfcl.web_search.no_snippet_accuracy": 18.0,
-        "bfcl.memory.accuracy": 15.7,
-        "bfcl.memory.kv_accuracy": 13.55,
-        "bfcl.memory.vector_accuracy": 14.19,
-        "bfcl.memory.recursive_summarization_accuracy": 19.35,
-        "bfcl.relevance.relevance_detection_accuracy": 81.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 82.39,
-        "bfcl.format_sensitivity.max_delta": 15.5,
-        "bfcl.format_sensitivity.stddev": 3.75
-      }
-    },
-    {
-      "model_id": "qwen/qwen3-4b-instruct-2507-fc",
-      "name": "Qwen3-4B-Instruct-2507 (FC)",
-      "developer": "qwen",
-      "scores": {
-        "bfcl.overall.rank": 54.0,
-        "bfcl.overall.overall_accuracy": 35.68,
-        "bfcl.overall.total_cost_usd": 6.37,
-        "bfcl.overall.latency_mean_s": 7.61,
-        "bfcl.overall.latency_std_s": 20.36,
-        "bfcl.overall.latency_p95_s": 49.18,
-        "bfcl.non_live.ast_accuracy": 87.88,
-        "bfcl.non_live.simple_ast_accuracy": 75.5,
-        "bfcl.non_live.multiple_ast_accuracy": 93.5,
-        "bfcl.non_live.parallel_ast_accuracy": 92.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 90.0,
-        "bfcl.live.live_accuracy": 76.39,
-        "bfcl.live.live_simple_ast_accuracy": 79.07,
-        "bfcl.live.live_multiple_ast_accuracy": 76.16,
-        "bfcl.live.live_parallel_ast_accuracy": 62.5,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 66.67,
-        "bfcl.multi_turn.accuracy": 22.12,
-        "bfcl.multi_turn.base_accuracy": 26.5,
-        "bfcl.multi_turn.miss_function_accuracy": 21.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 15.5,
-        "bfcl.multi_turn.long_context_accuracy": 25.5,
-        "bfcl.web_search.accuracy": 3.0,
-        "bfcl.web_search.base_accuracy": 4.0,
-        "bfcl.web_search.no_snippet_accuracy": 2.0,
-        "bfcl.memory.accuracy": 17.63,
-        "bfcl.memory.kv_accuracy": 16.13,
-        "bfcl.memory.vector_accuracy": 12.26,
-        "bfcl.memory.recursive_summarization_accuracy": 24.52,
-        "bfcl.relevance.relevance_detection_accuracy": 87.5,
-        "bfcl.relevance.irrelevance_detection_accuracy": 84.93
-      }
-    },
-    {
-      "model_id": "qwen/qwen3-4b-instruct-2507-prompt",
-      "name": "Qwen3-4B-Instruct-2507 (Prompt)",
-      "developer": "qwen",
-      "scores": {
-        "bfcl.overall.rank": 55.0,
-        "bfcl.overall.overall_accuracy": 35.52,
-        "bfcl.overall.total_cost_usd": 53.66,
-        "bfcl.overall.latency_mean_s": 44.7,
-        "bfcl.overall.latency_std_s": 163.79,
-        "bfcl.overall.latency_p95_s": 208.06,
-        "bfcl.non_live.ast_accuracy": 86.44,
-        "bfcl.non_live.simple_ast_accuracy": 77.25,
-        "bfcl.non_live.multiple_ast_accuracy": 91.0,
-        "bfcl.non_live.parallel_ast_accuracy": 88.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 89.5,
-        "bfcl.live.live_accuracy": 74.69,
-        "bfcl.live.live_simple_ast_accuracy": 77.91,
-        "bfcl.live.live_multiple_ast_accuracy": 74.17,
-        "bfcl.live.live_parallel_ast_accuracy": 81.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 58.33,
-        "bfcl.multi_turn.accuracy": 20.5,
-        "bfcl.multi_turn.base_accuracy": 24.5,
-        "bfcl.multi_turn.miss_function_accuracy": 21.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 16.0,
-        "bfcl.multi_turn.long_context_accuracy": 20.0,
-        "bfcl.web_search.accuracy": 4.5,
-        "bfcl.web_search.base_accuracy": 4.0,
-        "bfcl.web_search.no_snippet_accuracy": 5.0,
-        "bfcl.memory.accuracy": 23.87,
-        "bfcl.memory.kv_accuracy": 12.9,
-        "bfcl.memory.vector_accuracy": 14.19,
-        "bfcl.memory.recursive_summarization_accuracy": 44.52,
-        "bfcl.relevance.relevance_detection_accuracy": 87.5,
-        "bfcl.relevance.irrelevance_detection_accuracy": 75.87,
-        "bfcl.format_sensitivity.max_delta": 18.0,
-        "bfcl.format_sensitivity.stddev": 5.22
-      }
-    },
-    {
-      "model_id": "qwen/qwen3-8b-fc",
-      "name": "Qwen3-8B (FC)",
-      "developer": "qwen",
-      "scores": {
-        "bfcl.overall.rank": 39.0,
-        "bfcl.overall.overall_accuracy": 42.57,
-        "bfcl.overall.total_cost_usd": 43.32,
-        "bfcl.overall.latency_mean_s": 51.36,
-        "bfcl.overall.latency_std_s": 76.14,
-        "bfcl.overall.latency_p95_s": 188.98,
-        "bfcl.non_live.ast_accuracy": 87.58,
-        "bfcl.non_live.simple_ast_accuracy": 72.83,
-        "bfcl.non_live.multiple_ast_accuracy": 96.5,
-        "bfcl.non_live.parallel_ast_accuracy": 92.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 89.0,
-        "bfcl.live.live_accuracy": 80.53,
-        "bfcl.live.live_simple_ast_accuracy": 84.5,
-        "bfcl.live.live_multiple_ast_accuracy": 79.68,
-        "bfcl.live.live_parallel_ast_accuracy": 75.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 79.17,
-        "bfcl.multi_turn.accuracy": 41.75,
-        "bfcl.multi_turn.base_accuracy": 50.5,
-        "bfcl.multi_turn.miss_function_accuracy": 42.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 40.0,
-        "bfcl.multi_turn.long_context_accuracy": 34.5,
-        "bfcl.web_search.accuracy": 12.0,
-        "bfcl.web_search.base_accuracy": 15.0,
-        "bfcl.web_search.no_snippet_accuracy": 9.0,
-        "bfcl.memory.accuracy": 14.62,
-        "bfcl.memory.kv_accuracy": 5.16,
-        "bfcl.memory.vector_accuracy": 7.1,
-        "bfcl.memory.recursive_summarization_accuracy": 31.61,
-        "bfcl.relevance.relevance_detection_accuracy": 93.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 79.07
-      }
-    },
-    {
-      "model_id": "qwen/qwen3-8b-prompt",
-      "name": "Qwen3-8B (Prompt)",
-      "developer": "qwen",
-      "scores": {
-        "bfcl.overall.rank": 44.0,
-        "bfcl.overall.overall_accuracy": 40.43,
-        "bfcl.overall.total_cost_usd": 63.95,
-        "bfcl.overall.latency_mean_s": 54.17,
-        "bfcl.overall.latency_std_s": 79.9,
-        "bfcl.overall.latency_p95_s": 194.15,
-        "bfcl.non_live.ast_accuracy": 88.56,
-        "bfcl.non_live.simple_ast_accuracy": 75.25,
-        "bfcl.non_live.multiple_ast_accuracy": 95.0,
-        "bfcl.non_live.parallel_ast_accuracy": 94.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 89.5,
-        "bfcl.live.live_accuracy": 80.09,
-        "bfcl.live.live_simple_ast_accuracy": 84.5,
-        "bfcl.live.live_multiple_ast_accuracy": 78.92,
-        "bfcl.live.live_parallel_ast_accuracy": 93.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 75.0,
-        "bfcl.multi_turn.accuracy": 33.38,
-        "bfcl.multi_turn.base_accuracy": 41.5,
-        "bfcl.multi_turn.miss_function_accuracy": 38.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 27.0,
-        "bfcl.multi_turn.long_context_accuracy": 26.5,
-        "bfcl.web_search.accuracy": 13.5,
-        "bfcl.web_search.base_accuracy": 19.0,
-        "bfcl.web_search.no_snippet_accuracy": 8.0,
-        "bfcl.memory.accuracy": 13.12,
-        "bfcl.memory.kv_accuracy": 3.87,
-        "bfcl.memory.vector_accuracy": 10.32,
-        "bfcl.memory.recursive_summarization_accuracy": 25.16,
-        "bfcl.relevance.relevance_detection_accuracy": 75.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 82.27,
-        "bfcl.format_sensitivity.max_delta": 16.5,
-        "bfcl.format_sensitivity.stddev": 5.09
-      }
-    },
-    {
-      "model_id": "salesforce/xlam-2-1b-fc-r-fc",
-      "name": "xLAM-2-1b-fc-r (FC)",
-      "developer": "salesforce",
-      "scores": {
-        "bfcl.overall.rank": 65.0,
-        "bfcl.overall.overall_accuracy": 30.44,
-        "bfcl.overall.total_cost_usd": 2.79,
-        "bfcl.overall.latency_mean_s": 2.84,
-        "bfcl.overall.latency_std_s": 2.35,
-        "bfcl.overall.latency_p95_s": 6.52,
-        "bfcl.non_live.ast_accuracy": 69.04,
-        "bfcl.non_live.simple_ast_accuracy": 64.17,
-        "bfcl.non_live.multiple_ast_accuracy": 82.5,
-        "bfcl.non_live.parallel_ast_accuracy": 73.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 56.0,
-        "bfcl.live.live_accuracy": 55.14,
-        "bfcl.live.live_simple_ast_accuracy": 68.22,
-        "bfcl.live.live_multiple_ast_accuracy": 52.8,
-        "bfcl.live.live_parallel_ast_accuracy": 43.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 25.0,
-        "bfcl.multi_turn.accuracy": 36.0,
-        "bfcl.multi_turn.base_accuracy": 45.5,
-        "bfcl.multi_turn.miss_function_accuracy": 36.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 37.0,
-        "bfcl.multi_turn.long_context_accuracy": 25.5,
-        "bfcl.web_search.accuracy": 0.0,
-        "bfcl.web_search.base_accuracy": 0.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 3.87,
-        "bfcl.memory.kv_accuracy": 3.87,
-        "bfcl.memory.vector_accuracy": 3.87,
-        "bfcl.memory.recursive_summarization_accuracy": 3.87,
-        "bfcl.relevance.relevance_detection_accuracy": 87.5,
-        "bfcl.relevance.irrelevance_detection_accuracy": 64.47
-      }
-    },
-    {
-      "model_id": "salesforce/xlam-2-32b-fc-r-fc",
-      "name": "xLAM-2-32b-fc-r (FC)",
-      "developer": "salesforce",
-      "scores": {
-        "bfcl.overall.rank": 18.0,
-        "bfcl.overall.overall_accuracy": 54.66,
-        "bfcl.overall.total_cost_usd": 6.0,
-        "bfcl.overall.latency_mean_s": 6.94,
-        "bfcl.overall.latency_std_s": 8.21,
-        "bfcl.overall.latency_p95_s": 17.66,
-        "bfcl.non_live.ast_accuracy": 89.6,
-        "bfcl.non_live.simple_ast_accuracy": 80.42,
-        "bfcl.non_live.multiple_ast_accuracy": 94.0,
-        "bfcl.non_live.parallel_ast_accuracy": 93.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 91.0,
-        "bfcl.live.live_accuracy": 75.5,
-        "bfcl.live.live_simple_ast_accuracy": 82.17,
-        "bfcl.live.live_multiple_ast_accuracy": 74.64,
-        "bfcl.live.live_parallel_ast_accuracy": 50.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 58.33,
-        "bfcl.multi_turn.accuracy": 69.5,
-        "bfcl.multi_turn.base_accuracy": 81.5,
-        "bfcl.multi_turn.miss_function_accuracy": 72.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 67.5,
-        "bfcl.multi_turn.long_context_accuracy": 56.5,
-        "bfcl.web_search.accuracy": 25.5,
-        "bfcl.web_search.base_accuracy": 37.0,
-        "bfcl.web_search.no_snippet_accuracy": 14.0,
-        "bfcl.memory.accuracy": 20.86,
-        "bfcl.memory.kv_accuracy": 6.45,
-        "bfcl.memory.vector_accuracy": 10.32,
-        "bfcl.memory.recursive_summarization_accuracy": 45.81,
-        "bfcl.relevance.relevance_detection_accuracy": 81.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 80.23
-      }
-    },
-    {
-      "model_id": "salesforce/xlam-2-3b-fc-r-fc",
-      "name": "xLAM-2-3b-fc-r (FC)",
-      "developer": "salesforce",
-      "scores": {
-        "bfcl.overall.rank": 42.0,
-        "bfcl.overall.overall_accuracy": 41.22,
-        "bfcl.overall.total_cost_usd": 3.36,
-        "bfcl.overall.latency_mean_s": 3.8,
-        "bfcl.overall.latency_std_s": 3.59,
-        "bfcl.overall.latency_p95_s": 8.79,
-        "bfcl.non_live.ast_accuracy": 82.96,
-        "bfcl.non_live.simple_ast_accuracy": 75.33,
-        "bfcl.non_live.multiple_ast_accuracy": 91.0,
-        "bfcl.non_live.parallel_ast_accuracy": 86.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 79.0,
-        "bfcl.live.live_accuracy": 62.92,
-        "bfcl.live.live_simple_ast_accuracy": 73.26,
-        "bfcl.live.live_multiple_ast_accuracy": 60.68,
-        "bfcl.live.live_parallel_ast_accuracy": 62.5,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 50.0,
-        "bfcl.multi_turn.accuracy": 58.38,
-        "bfcl.multi_turn.base_accuracy": 71.5,
-        "bfcl.multi_turn.miss_function_accuracy": 59.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 57.5,
-        "bfcl.multi_turn.long_context_accuracy": 45.5,
-        "bfcl.web_search.accuracy": 2.5,
-        "bfcl.web_search.base_accuracy": 3.0,
-        "bfcl.web_search.no_snippet_accuracy": 2.0,
-        "bfcl.memory.accuracy": 11.4,
-        "bfcl.memory.kv_accuracy": 5.81,
-        "bfcl.memory.vector_accuracy": 5.81,
-        "bfcl.memory.recursive_summarization_accuracy": 22.58,
-        "bfcl.relevance.relevance_detection_accuracy": 87.5,
-        "bfcl.relevance.irrelevance_detection_accuracy": 63.45
-      }
-    },
-    {
-      "model_id": "salesforce/xlam-2-70b-fc-r-fc",
-      "name": "xLAM-2-70b-fc-r (FC)",
-      "developer": "salesforce",
-      "scores": {
-        "bfcl.overall.rank": 22.0,
-        "bfcl.overall.overall_accuracy": 53.07,
-        "bfcl.overall.total_cost_usd": 25.1,
-        "bfcl.overall.latency_mean_s": 28.06,
-        "bfcl.overall.latency_std_s": 68.77,
-        "bfcl.overall.latency_p95_s": 91.21,
-        "bfcl.non_live.ast_accuracy": 88.44,
-        "bfcl.non_live.simple_ast_accuracy": 78.25,
-        "bfcl.non_live.multiple_ast_accuracy": 94.0,
-        "bfcl.non_live.parallel_ast_accuracy": 92.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 89.5,
-        "bfcl.live.live_accuracy": 72.17,
-        "bfcl.live.live_simple_ast_accuracy": 77.91,
-        "bfcl.live.live_multiple_ast_accuracy": 71.13,
-        "bfcl.live.live_parallel_ast_accuracy": 68.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 58.33,
-        "bfcl.multi_turn.accuracy": 77.38,
-        "bfcl.multi_turn.base_accuracy": 82.5,
-        "bfcl.multi_turn.miss_function_accuracy": 77.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 74.0,
-        "bfcl.multi_turn.long_context_accuracy": 76.0,
-        "bfcl.web_search.accuracy": 15.0,
-        "bfcl.web_search.base_accuracy": 17.0,
-        "bfcl.web_search.no_snippet_accuracy": 13.0,
-        "bfcl.memory.accuracy": 14.41,
-        "bfcl.memory.kv_accuracy": 2.58,
-        "bfcl.memory.vector_accuracy": 10.97,
-        "bfcl.memory.recursive_summarization_accuracy": 29.68,
-        "bfcl.relevance.relevance_detection_accuracy": 75.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 79.11
-      }
-    },
-    {
-      "model_id": "salesforce/xlam-2-8b-fc-r-fc",
-      "name": "xLAM-2-8b-fc-r (FC)",
-      "developer": "salesforce",
-      "scores": {
-        "bfcl.overall.rank": 34.0,
-        "bfcl.overall.overall_accuracy": 46.68,
-        "bfcl.overall.total_cost_usd": 20.92,
-        "bfcl.overall.latency_mean_s": 22.65,
-        "bfcl.overall.latency_std_s": 46.92,
-        "bfcl.overall.latency_p95_s": 108.81,
-        "bfcl.non_live.ast_accuracy": 84.58,
-        "bfcl.non_live.simple_ast_accuracy": 73.83,
-        "bfcl.non_live.multiple_ast_accuracy": 93.5,
-        "bfcl.non_live.parallel_ast_accuracy": 87.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 83.5,
-        "bfcl.live.live_accuracy": 67.95,
-        "bfcl.live.live_simple_ast_accuracy": 75.58,
-        "bfcl.live.live_multiple_ast_accuracy": 66.57,
-        "bfcl.live.live_parallel_ast_accuracy": 56.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 54.17,
-        "bfcl.multi_turn.accuracy": 70.0,
-        "bfcl.multi_turn.base_accuracy": 76.0,
-        "bfcl.multi_turn.miss_function_accuracy": 72.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 65.0,
-        "bfcl.multi_turn.long_context_accuracy": 67.0,
-        "bfcl.web_search.accuracy": 6.5,
-        "bfcl.web_search.base_accuracy": 11.0,
-        "bfcl.web_search.no_snippet_accuracy": 2.0,
-        "bfcl.memory.accuracy": 13.98,
-        "bfcl.memory.kv_accuracy": 5.81,
-        "bfcl.memory.vector_accuracy": 15.48,
-        "bfcl.memory.recursive_summarization_accuracy": 20.65,
-        "bfcl.relevance.relevance_detection_accuracy": 87.5,
-        "bfcl.relevance.irrelevance_detection_accuracy": 63.28
-      }
-    },
-    {
-      "model_id": "speakleash-ack-cyfronet-agh/bielik-11b-v2-3-instruct-prompt",
-      "name": "Bielik-11B-v2.3-Instruct (Prompt)",
-      "developer": "speakleash-ack-cyfronet-agh",
-      "scores": {
-        "bfcl.overall.rank": 99.0,
-        "bfcl.overall.overall_accuracy": 21.9,
-        "bfcl.overall.total_cost_usd": 22.44,
-        "bfcl.overall.latency_mean_s": 23.75,
-        "bfcl.overall.latency_std_s": 61.76,
-        "bfcl.overall.latency_p95_s": 72.8,
-        "bfcl.non_live.ast_accuracy": 81.5,
-        "bfcl.non_live.simple_ast_accuracy": 73.0,
-        "bfcl.non_live.multiple_ast_accuracy": 92.0,
-        "bfcl.non_live.parallel_ast_accuracy": 85.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 75.5,
-        "bfcl.live.live_accuracy": 67.8,
-        "bfcl.live.live_simple_ast_accuracy": 75.58,
-        "bfcl.live.live_multiple_ast_accuracy": 66.19,
-        "bfcl.live.live_parallel_ast_accuracy": 62.5,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 58.33,
-        "bfcl.multi_turn.accuracy": 2.62,
-        "bfcl.multi_turn.base_accuracy": 4.5,
-        "bfcl.multi_turn.miss_function_accuracy": 0.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 3.0,
-        "bfcl.multi_turn.long_context_accuracy": 2.5,
-        "bfcl.web_search.accuracy": 1.5,
-        "bfcl.web_search.base_accuracy": 1.0,
-        "bfcl.web_search.no_snippet_accuracy": 2.0,
-        "bfcl.memory.accuracy": 11.4,
-        "bfcl.memory.kv_accuracy": 7.1,
-        "bfcl.memory.vector_accuracy": 4.52,
-        "bfcl.memory.recursive_summarization_accuracy": 22.58,
-        "bfcl.relevance.relevance_detection_accuracy": 93.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 36.01,
-        "bfcl.format_sensitivity.max_delta": 35.0,
-        "bfcl.format_sensitivity.stddev": 9.74
-      }
-    },
-    {
-      "model_id": "tii-uae/falcon3-10b-instruct-fc",
-      "name": "Falcon3-10B-Instruct (FC)",
-      "developer": "tii-uae",
-      "scores": {
-        "bfcl.overall.rank": 82.0,
-        "bfcl.overall.overall_accuracy": 27.01,
-        "bfcl.overall.total_cost_usd": 52.59,
-        "bfcl.overall.latency_mean_s": 69.27,
-        "bfcl.overall.latency_std_s": 92.22,
-        "bfcl.overall.latency_p95_s": 190.96,
-        "bfcl.non_live.ast_accuracy": 85.0,
-        "bfcl.non_live.simple_ast_accuracy": 70.5,
-        "bfcl.non_live.multiple_ast_accuracy": 93.5,
-        "bfcl.non_live.parallel_ast_accuracy": 88.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 87.5,
-        "bfcl.live.live_accuracy": 75.43,
-        "bfcl.live.live_simple_ast_accuracy": 77.13,
-        "bfcl.live.live_multiple_ast_accuracy": 76.16,
-        "bfcl.live.live_parallel_ast_accuracy": 50.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 41.67,
-        "bfcl.multi_turn.accuracy": 6.5,
-        "bfcl.multi_turn.base_accuracy": 6.5,
-        "bfcl.multi_turn.miss_function_accuracy": 9.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 5.0,
-        "bfcl.multi_turn.long_context_accuracy": 5.0,
-        "bfcl.web_search.accuracy": 1.5,
-        "bfcl.web_search.base_accuracy": 2.0,
-        "bfcl.web_search.no_snippet_accuracy": 1.0,
-        "bfcl.memory.accuracy": 27.53,
-        "bfcl.memory.kv_accuracy": 12.26,
-        "bfcl.memory.vector_accuracy": 19.35,
-        "bfcl.memory.recursive_summarization_accuracy": 50.97,
-        "bfcl.relevance.relevance_detection_accuracy": 93.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 32.09
-      }
-    },
-    {
-      "model_id": "tii-uae/falcon3-1b-instruct-fc",
-      "name": "Falcon3-1B-Instruct (FC)",
-      "developer": "tii-uae",
-      "scores": {
-        "bfcl.overall.rank": 106.0,
-        "bfcl.overall.overall_accuracy": 11.08,
-        "bfcl.overall.total_cost_usd": 1.72,
-        "bfcl.overall.latency_mean_s": 5.23,
-        "bfcl.overall.latency_std_s": 14.34,
-        "bfcl.overall.latency_p95_s": 11.48,
-        "bfcl.non_live.ast_accuracy": 9.02,
-        "bfcl.non_live.simple_ast_accuracy": 2.58,
-        "bfcl.non_live.multiple_ast_accuracy": 6.0,
-        "bfcl.non_live.parallel_ast_accuracy": 18.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 9.5,
-        "bfcl.live.live_accuracy": 2.89,
-        "bfcl.live.live_simple_ast_accuracy": 4.26,
-        "bfcl.live.live_multiple_ast_accuracy": 2.37,
-        "bfcl.live.live_parallel_ast_accuracy": 0.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 12.5,
-        "bfcl.multi_turn.accuracy": 0.0,
-        "bfcl.multi_turn.base_accuracy": 0.0,
-        "bfcl.multi_turn.miss_function_accuracy": 0.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 0.0,
-        "bfcl.multi_turn.long_context_accuracy": 0.0,
-        "bfcl.web_search.accuracy": 0.0,
-        "bfcl.web_search.base_accuracy": 0.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 5.81,
-        "bfcl.memory.kv_accuracy": 5.16,
-        "bfcl.memory.vector_accuracy": 7.74,
-        "bfcl.memory.recursive_summarization_accuracy": 4.52,
-        "bfcl.relevance.relevance_detection_accuracy": 0.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 87.3
-      }
-    },
-    {
-      "model_id": "tii-uae/falcon3-3b-instruct-fc",
-      "name": "Falcon3-3B-Instruct (FC)",
-      "developer": "tii-uae",
-      "scores": {
-        "bfcl.overall.rank": 104.0,
-        "bfcl.overall.overall_accuracy": 16.25,
-        "bfcl.overall.total_cost_usd": 36.7,
-        "bfcl.overall.latency_mean_s": 38.52,
-        "bfcl.overall.latency_std_s": 107.47,
-        "bfcl.overall.latency_p95_s": 103.62,
-        "bfcl.non_live.ast_accuracy": 54.62,
-        "bfcl.non_live.simple_ast_accuracy": 56.5,
-        "bfcl.non_live.multiple_ast_accuracy": 69.5,
-        "bfcl.non_live.parallel_ast_accuracy": 67.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 25.5,
-        "bfcl.live.live_accuracy": 54.48,
-        "bfcl.live.live_simple_ast_accuracy": 57.36,
-        "bfcl.live.live_multiple_ast_accuracy": 54.7,
-        "bfcl.live.live_parallel_ast_accuracy": 25.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 33.33,
-        "bfcl.multi_turn.accuracy": 1.0,
-        "bfcl.multi_turn.base_accuracy": 1.5,
-        "bfcl.multi_turn.miss_function_accuracy": 0.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 0.5,
-        "bfcl.multi_turn.long_context_accuracy": 1.5,
-        "bfcl.web_search.accuracy": 1.0,
-        "bfcl.web_search.base_accuracy": 1.0,
-        "bfcl.web_search.no_snippet_accuracy": 1.0,
-        "bfcl.memory.accuracy": 7.74,
-        "bfcl.memory.kv_accuracy": 6.45,
-        "bfcl.memory.vector_accuracy": 8.39,
-        "bfcl.memory.recursive_summarization_accuracy": 8.39,
-        "bfcl.relevance.relevance_detection_accuracy": 81.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 32.92
-      }
-    },
-    {
-      "model_id": "tii-uae/falcon3-7b-instruct-fc",
-      "name": "Falcon3-7B-Instruct (FC)",
-      "developer": "tii-uae",
-      "scores": {
-        "bfcl.overall.rank": 91.0,
-        "bfcl.overall.overall_accuracy": 24.03,
-        "bfcl.overall.total_cost_usd": 73.61,
-        "bfcl.overall.latency_mean_s": 93.11,
-        "bfcl.overall.latency_std_s": 117.8,
-        "bfcl.overall.latency_p95_s": 315.7,
-        "bfcl.non_live.ast_accuracy": 82.69,
-        "bfcl.non_live.simple_ast_accuracy": 65.75,
-        "bfcl.non_live.multiple_ast_accuracy": 89.0,
-        "bfcl.non_live.parallel_ast_accuracy": 87.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 89.0,
-        "bfcl.live.live_accuracy": 68.32,
-        "bfcl.live.live_simple_ast_accuracy": 74.81,
-        "bfcl.live.live_multiple_ast_accuracy": 66.76,
-        "bfcl.live.live_parallel_ast_accuracy": 75.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 62.5,
-        "bfcl.multi_turn.accuracy": 5.0,
-        "bfcl.multi_turn.base_accuracy": 7.0,
-        "bfcl.multi_turn.miss_function_accuracy": 4.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 5.0,
-        "bfcl.multi_turn.long_context_accuracy": 4.0,
-        "bfcl.web_search.accuracy": 0.5,
-        "bfcl.web_search.base_accuracy": 1.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 20.65,
-        "bfcl.memory.kv_accuracy": 10.32,
-        "bfcl.memory.vector_accuracy": 12.9,
-        "bfcl.memory.recursive_summarization_accuracy": 38.71,
-        "bfcl.relevance.relevance_detection_accuracy": 100.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 31.99
-      }
-    },
-    {
-      "model_id": "uiuc-oumi/coalm-70b",
-      "name": "CoALM-70B",
-      "developer": "uiuc-oumi",
-      "scores": {
-        "bfcl.overall.rank": 74.0,
-        "bfcl.overall.overall_accuracy": 27.99,
-        "bfcl.overall.total_cost_usd": 19.89,
-        "bfcl.overall.latency_mean_s": 16.22,
-        "bfcl.overall.latency_std_s": 59.91,
-        "bfcl.overall.latency_p95_s": 36.0,
-        "bfcl.non_live.ast_accuracy": 83.44,
-        "bfcl.non_live.simple_ast_accuracy": 70.25,
-        "bfcl.non_live.multiple_ast_accuracy": 92.0,
-        "bfcl.non_live.parallel_ast_accuracy": 88.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 83.0,
-        "bfcl.live.live_accuracy": 67.28,
-        "bfcl.live.live_simple_ast_accuracy": 70.54,
-        "bfcl.live.live_multiple_ast_accuracy": 66.57,
-        "bfcl.live.live_parallel_ast_accuracy": 68.75,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 62.5,
-        "bfcl.multi_turn.accuracy": 10.62,
-        "bfcl.multi_turn.base_accuracy": 11.0,
-        "bfcl.multi_turn.miss_function_accuracy": 14.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 9.0,
-        "bfcl.multi_turn.long_context_accuracy": 8.5,
-        "bfcl.web_search.accuracy": 0.0,
-        "bfcl.web_search.base_accuracy": 0.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 5.81,
-        "bfcl.memory.kv_accuracy": 9.03,
-        "bfcl.memory.vector_accuracy": 5.16,
-        "bfcl.memory.recursive_summarization_accuracy": 3.23,
-        "bfcl.relevance.relevance_detection_accuracy": 93.75,
-        "bfcl.relevance.irrelevance_detection_accuracy": 85.65,
-        "bfcl.format_sensitivity.max_delta": 72.0,
-        "bfcl.format_sensitivity.stddev": 27.76
-      }
-    },
-    {
-      "model_id": "uiuc-oumi/coalm-8b",
-      "name": "CoALM-8B",
-      "developer": "uiuc-oumi",
-      "scores": {
-        "bfcl.overall.rank": 84.0,
-        "bfcl.overall.overall_accuracy": 26.81,
-        "bfcl.overall.total_cost_usd": 25.33,
-        "bfcl.overall.latency_mean_s": 20.36,
-        "bfcl.overall.latency_std_s": 73.74,
-        "bfcl.overall.latency_p95_s": 138.04,
-        "bfcl.non_live.ast_accuracy": 84.87,
-        "bfcl.non_live.simple_ast_accuracy": 69.5,
-        "bfcl.non_live.multiple_ast_accuracy": 93.5,
-        "bfcl.non_live.parallel_ast_accuracy": 88.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 88.5,
-        "bfcl.live.live_accuracy": 66.77,
-        "bfcl.live.live_simple_ast_accuracy": 70.54,
-        "bfcl.live.live_multiple_ast_accuracy": 66.19,
-        "bfcl.live.live_parallel_ast_accuracy": 62.5,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 54.17,
-        "bfcl.multi_turn.accuracy": 8.0,
-        "bfcl.multi_turn.base_accuracy": 10.0,
-        "bfcl.multi_turn.miss_function_accuracy": 7.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 8.0,
-        "bfcl.multi_turn.long_context_accuracy": 7.0,
-        "bfcl.web_search.accuracy": 0.0,
-        "bfcl.web_search.base_accuracy": 0.0,
-        "bfcl.web_search.no_snippet_accuracy": 0.0,
-        "bfcl.memory.accuracy": 2.8,
-        "bfcl.memory.kv_accuracy": 3.23,
-        "bfcl.memory.vector_accuracy": 3.87,
-        "bfcl.memory.recursive_summarization_accuracy": 1.29,
-        "bfcl.relevance.relevance_detection_accuracy": 87.5,
-        "bfcl.relevance.irrelevance_detection_accuracy": 86.9,
-        "bfcl.format_sensitivity.max_delta": 79.0,
-        "bfcl.format_sensitivity.stddev": 34.18
-      }
-    },
-    {
-      "model_id": "writer/palmyra-x-004-fc",
-      "name": "palmyra-x-004 (FC)",
-      "developer": "writer",
-      "scores": {
-        "bfcl.overall.rank": 76.0,
-        "bfcl.overall.overall_accuracy": 27.87,
-        "bfcl.overall.total_cost_usd": 178.15,
-        "bfcl.overall.latency_mean_s": 3.71,
-        "bfcl.overall.latency_std_s": 7.62,
-        "bfcl.overall.latency_p95_s": 8.04,
-        "bfcl.non_live.ast_accuracy": 87.46,
-        "bfcl.non_live.simple_ast_accuracy": 71.33,
-        "bfcl.non_live.multiple_ast_accuracy": 96.0,
-        "bfcl.non_live.parallel_ast_accuracy": 93.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 89.5,
-        "bfcl.live.live_accuracy": 77.87,
-        "bfcl.live.live_simple_ast_accuracy": 79.46,
-        "bfcl.live.live_multiple_ast_accuracy": 77.97,
-        "bfcl.live.live_parallel_ast_accuracy": 56.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 70.83,
-        "bfcl.multi_turn.accuracy": 0.38,
-        "bfcl.multi_turn.base_accuracy": 0.5,
-        "bfcl.multi_turn.miss_function_accuracy": 0.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 0.5,
-        "bfcl.multi_turn.long_context_accuracy": 0.5,
-        "bfcl.web_search.accuracy": 2.5,
-        "bfcl.web_search.base_accuracy": 4.0,
-        "bfcl.web_search.no_snippet_accuracy": 1.0,
-        "bfcl.memory.accuracy": 13.12,
-        "bfcl.memory.kv_accuracy": 6.45,
-        "bfcl.memory.vector_accuracy": 14.19,
-        "bfcl.memory.recursive_summarization_accuracy": 18.71,
-        "bfcl.relevance.relevance_detection_accuracy": 81.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 80.99
-      }
-    },
-    {
-      "model_id": "xai/grok-4-0709-fc",
-      "name": "Grok-4-0709 (FC)",
-      "developer": "xAI",
-      "scores": {
-        "bfcl.overall.rank": 10.0,
-        "bfcl.overall.overall_accuracy": 61.38,
-        "bfcl.overall.total_cost_usd": 355.17,
-        "bfcl.overall.latency_mean_s": 15.49,
-        "bfcl.overall.latency_std_s": 26.22,
-        "bfcl.overall.latency_p95_s": 44.28,
-        "bfcl.non_live.ast_accuracy": 85.38,
-        "bfcl.non_live.simple_ast_accuracy": 73.5,
-        "bfcl.non_live.multiple_ast_accuracy": 92.5,
-        "bfcl.non_live.parallel_ast_accuracy": 88.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 87.0,
-        "bfcl.live.live_accuracy": 75.57,
-        "bfcl.live.live_simple_ast_accuracy": 82.17,
-        "bfcl.live.live_multiple_ast_accuracy": 73.88,
-        "bfcl.live.live_parallel_ast_accuracy": 75.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 79.17,
-        "bfcl.multi_turn.accuracy": 33.88,
-        "bfcl.multi_turn.base_accuracy": 44.0,
-        "bfcl.multi_turn.miss_function_accuracy": 19.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 28.5,
-        "bfcl.multi_turn.long_context_accuracy": 44.0,
-        "bfcl.web_search.accuracy": 82.0,
-        "bfcl.web_search.base_accuracy": 80.0,
-        "bfcl.web_search.no_snippet_accuracy": 84.0,
-        "bfcl.memory.accuracy": 55.91,
-        "bfcl.memory.kv_accuracy": 57.42,
-        "bfcl.memory.vector_accuracy": 58.71,
-        "bfcl.memory.recursive_summarization_accuracy": 51.61,
-        "bfcl.relevance.relevance_detection_accuracy": 87.5,
-        "bfcl.relevance.irrelevance_detection_accuracy": 75.4
-      }
-    },
-    {
-      "model_id": "xai/grok-4-0709-prompt",
-      "name": "Grok-4-0709 (Prompt)",
-      "developer": "xAI",
-      "scores": {
-        "bfcl.overall.rank": 9.0,
-        "bfcl.overall.overall_accuracy": 62.97,
-        "bfcl.overall.total_cost_usd": 348.19,
-        "bfcl.overall.latency_mean_s": 30.38,
-        "bfcl.overall.latency_std_s": 36.19,
-        "bfcl.overall.latency_p95_s": 101.54,
-        "bfcl.non_live.ast_accuracy": 82.75,
-        "bfcl.non_live.simple_ast_accuracy": 67.0,
-        "bfcl.non_live.multiple_ast_accuracy": 93.5,
-        "bfcl.non_live.parallel_ast_accuracy": 89.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 81.5,
-        "bfcl.live.live_accuracy": 72.54,
-        "bfcl.live.live_simple_ast_accuracy": 81.78,
-        "bfcl.live.live_multiple_ast_accuracy": 70.18,
-        "bfcl.live.live_parallel_ast_accuracy": 81.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 70.83,
-        "bfcl.multi_turn.accuracy": 47.0,
-        "bfcl.multi_turn.base_accuracy": 55.5,
-        "bfcl.multi_turn.miss_function_accuracy": 46.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 36.0,
-        "bfcl.multi_turn.long_context_accuracy": 50.5,
-        "bfcl.web_search.accuracy": 74.0,
-        "bfcl.web_search.base_accuracy": 74.0,
-        "bfcl.web_search.no_snippet_accuracy": 74.0,
-        "bfcl.memory.accuracy": 50.54,
-        "bfcl.memory.kv_accuracy": 43.87,
-        "bfcl.memory.vector_accuracy": 59.35,
-        "bfcl.memory.recursive_summarization_accuracy": 48.39,
-        "bfcl.relevance.relevance_detection_accuracy": 81.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 84.3,
-        "bfcl.format_sensitivity.max_delta": 13.0,
-        "bfcl.format_sensitivity.stddev": 2.88
-      }
-    },
-    {
-      "model_id": "xai/grok-4-1-fast-non-reasoning-fc",
-      "name": "Grok-4-1-fast-non-reasoning (FC)",
-      "developer": "xAI",
-      "scores": {
-        "bfcl.overall.rank": 12.0,
-        "bfcl.overall.overall_accuracy": 58.29,
-        "bfcl.overall.total_cost_usd": 16.27,
-        "bfcl.overall.latency_mean_s": 2.29,
-        "bfcl.overall.latency_std_s": 7.31,
-        "bfcl.overall.latency_p95_s": 5.34,
-        "bfcl.non_live.ast_accuracy": 88.13,
-        "bfcl.non_live.simple_ast_accuracy": 76.0,
-        "bfcl.non_live.multiple_ast_accuracy": 93.0,
-        "bfcl.non_live.parallel_ast_accuracy": 93.0,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 90.5,
-        "bfcl.live.live_accuracy": 77.94,
-        "bfcl.live.live_simple_ast_accuracy": 82.95,
-        "bfcl.live.live_multiple_ast_accuracy": 76.92,
-        "bfcl.live.live_parallel_ast_accuracy": 75.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 70.83,
-        "bfcl.multi_turn.accuracy": 46.75,
-        "bfcl.multi_turn.base_accuracy": 58.0,
-        "bfcl.multi_turn.miss_function_accuracy": 39.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 37.5,
-        "bfcl.multi_turn.long_context_accuracy": 52.0,
-        "bfcl.web_search.accuracy": 75.0,
-        "bfcl.web_search.base_accuracy": 74.0,
-        "bfcl.web_search.no_snippet_accuracy": 76.0,
-        "bfcl.memory.accuracy": 26.24,
-        "bfcl.memory.kv_accuracy": 20.65,
-        "bfcl.memory.vector_accuracy": 20.0,
-        "bfcl.memory.recursive_summarization_accuracy": 38.06,
-        "bfcl.relevance.relevance_detection_accuracy": 81.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 74.09
-      }
-    },
-    {
-      "model_id": "xai/grok-4-1-fast-reasoning-fc",
-      "name": "Grok-4-1-fast-reasoning (FC)",
-      "developer": "xAI",
-      "scores": {
-        "bfcl.overall.rank": 5.0,
-        "bfcl.overall.overall_accuracy": 69.57,
-        "bfcl.overall.total_cost_usd": 17.26,
-        "bfcl.overall.latency_mean_s": 6.74,
-        "bfcl.overall.latency_std_s": 12.78,
-        "bfcl.overall.latency_p95_s": 17.57,
-        "bfcl.non_live.ast_accuracy": 88.27,
-        "bfcl.non_live.simple_ast_accuracy": 77.58,
-        "bfcl.non_live.multiple_ast_accuracy": 93.0,
-        "bfcl.non_live.parallel_ast_accuracy": 92.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 90.0,
-        "bfcl.live.live_accuracy": 78.46,
-        "bfcl.live.live_simple_ast_accuracy": 84.11,
-        "bfcl.live.live_multiple_ast_accuracy": 77.3,
-        "bfcl.live.live_parallel_ast_accuracy": 75.0,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 70.83,
-        "bfcl.multi_turn.accuracy": 58.87,
-        "bfcl.multi_turn.base_accuracy": 70.5,
-        "bfcl.multi_turn.miss_function_accuracy": 59.5,
-        "bfcl.multi_turn.miss_parameter_accuracy": 43.0,
-        "bfcl.multi_turn.long_context_accuracy": 62.5,
-        "bfcl.web_search.accuracy": 82.5,
-        "bfcl.web_search.base_accuracy": 82.0,
-        "bfcl.web_search.no_snippet_accuracy": 83.0,
-        "bfcl.memory.accuracy": 53.98,
-        "bfcl.memory.kv_accuracy": 41.29,
-        "bfcl.memory.vector_accuracy": 57.42,
-        "bfcl.memory.recursive_summarization_accuracy": 63.23,
-        "bfcl.relevance.relevance_detection_accuracy": 81.25,
-        "bfcl.relevance.irrelevance_detection_accuracy": 79.43
-      }
-    },
-    {
-      "model_id": "zhipu/glm-4-6-fc-thinking",
-      "name": "GLM-4.6 (FC thinking)",
-      "developer": "zhipu",
-      "scores": {
-        "bfcl.overall.rank": 4.0,
-        "bfcl.overall.overall_accuracy": 72.38,
-        "bfcl.overall.total_cost_usd": 4.64,
-        "bfcl.overall.latency_mean_s": 4.34,
-        "bfcl.overall.latency_std_s": 7.22,
-        "bfcl.overall.latency_p95_s": 13.5,
-        "bfcl.non_live.ast_accuracy": 87.56,
-        "bfcl.non_live.simple_ast_accuracy": 74.25,
-        "bfcl.non_live.multiple_ast_accuracy": 95.0,
-        "bfcl.non_live.parallel_ast_accuracy": 91.5,
-        "bfcl.non_live.parallel_multiple_ast_accuracy": 89.5,
-        "bfcl.live.live_accuracy": 80.9,
-        "bfcl.live.live_simple_ast_accuracy": 89.53,
-        "bfcl.live.live_multiple_ast_accuracy": 78.92,
-        "bfcl.live.live_parallel_ast_accuracy": 81.25,
-        "bfcl.live.live_parallel_multiple_ast_accuracy": 75.0,
-        "bfcl.multi_turn.accuracy": 68.0,
-        "bfcl.multi_turn.base_accuracy": 74.5,
-        "bfcl.multi_turn.miss_function_accuracy": 68.0,
-        "bfcl.multi_turn.miss_parameter_accuracy": 63.0,
-        "bfcl.multi_turn.long_context_accuracy": 66.5,
-        "bfcl.web_search.accuracy": 77.5,
-        "bfcl.web_search.base_accuracy": 79.0,
-        "bfcl.web_search.no_snippet_accuracy": 76.0,
-        "bfcl.memory.accuracy": 55.7,
-        "bfcl.memory.kv_accuracy": 43.87,
-        "bfcl.memory.vector_accuracy": 56.13,
-        "bfcl.memory.recursive_summarization_accuracy": 67.1,
-        "bfcl.relevance.relevance_detection_accuracy": 75.0,
-        "bfcl.relevance.irrelevance_detection_accuracy": 84.96
-      }
-    }
-  ]
-}
\ No newline at end of file