Add pipeline_tag and improve usage snippets

#1
by nielsr HF Staff - opened
Files changed (1) hide show
  1. README.md +6 -3
README.md CHANGED
@@ -1,7 +1,10 @@
1
  ---
 
2
  language:
3
  - en
4
  license: apache-2.0
 
 
5
  tags:
6
  - quantization
7
  - sinq
@@ -11,7 +14,6 @@ tags:
11
  - qwen
12
  - llm
13
  - compression
14
- base_model: Qwen/Qwen3-14B
15
  base_model_relation: quantized
16
  ---
17
 
@@ -48,7 +50,7 @@ To support the project please put a star ⭐ in the official [SINQ](https://gith
48
 
49
  ---
50
 
51
- # 🚀 Usage</span>
52
 
53
  ## Prerequisite
54
  Before running the quantization script, make sure the **SINQ** library is installed.
@@ -58,6 +60,7 @@ Installation instructions and setup details are available in the [SINQ official
58
  You can load and use the model with our wrapper based on the 🤗 Transformers library:
59
 
60
  ```python
 
61
  from transformers import AutoTokenizer
62
  from sinq.patch_model import AutoSINQHFModel
63
 
@@ -74,7 +77,6 @@ inputs = tokenizer(prompt, return_tensors="pt").to("cuda:0")
74
  with torch.inference_mode():
75
  out_ids = sinq_model.generate(**inputs, max_new_tokens=32, do_sample=False)
76
  print(tokenizer.decode(out_ids[0], skip_special_tokens=True))
77
-
78
  ```
79
 
80
  <details>
@@ -83,6 +85,7 @@ print(tokenizer.decode(out_ids[0], skip_special_tokens=True))
83
  The quantized model was obtained using the **SINQ** quantization library, following the steps below:
84
 
85
  ```python
 
86
  from transformers import AutoModelForCausalLM, AutoTokenizer
87
  from sinq.patch_model import AutoSINQHFModel
88
  from sinq.sinqlinear import BaseQuantizeConfig
 
1
  ---
2
+ base_model: Qwen/Qwen3-14B
3
  language:
4
  - en
5
  license: apache-2.0
6
+ pipeline_tag: text-generation
7
+ library_name: sinq
8
  tags:
9
  - quantization
10
  - sinq
 
14
  - qwen
15
  - llm
16
  - compression
 
17
  base_model_relation: quantized
18
  ---
19
 
 
50
 
51
  ---
52
 
53
+ # 🚀 Usage
54
 
55
  ## Prerequisite
56
  Before running the quantization script, make sure the **SINQ** library is installed.
 
60
  You can load and use the model with our wrapper based on the 🤗 Transformers library:
61
 
62
  ```python
63
+ import torch
64
  from transformers import AutoTokenizer
65
  from sinq.patch_model import AutoSINQHFModel
66
 
 
77
  with torch.inference_mode():
78
  out_ids = sinq_model.generate(**inputs, max_new_tokens=32, do_sample=False)
79
  print(tokenizer.decode(out_ids[0], skip_special_tokens=True))
 
80
  ```
81
 
82
  <details>
 
85
  The quantized model was obtained using the **SINQ** quantization library, following the steps below:
86
 
87
  ```python
88
+ import torch
89
  from transformers import AutoModelForCausalLM, AutoTokenizer
90
  from sinq.patch_model import AutoSINQHFModel
91
  from sinq.sinqlinear import BaseQuantizeConfig