Instructions to use tellang/yeji-8b-rslora-v7-AWQ with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tellang/yeji-8b-rslora-v7-AWQ with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="tellang/yeji-8b-rslora-v7-AWQ")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("tellang/yeji-8b-rslora-v7-AWQ")
model = AutoModelForCausalLM.from_pretrained("tellang/yeji-8b-rslora-v7-AWQ")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use tellang/yeji-8b-rslora-v7-AWQ with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tellang/yeji-8b-rslora-v7-AWQ"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tellang/yeji-8b-rslora-v7-AWQ",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/tellang/yeji-8b-rslora-v7-AWQ

SGLang

How to use tellang/yeji-8b-rslora-v7-AWQ with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "tellang/yeji-8b-rslora-v7-AWQ" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tellang/yeji-8b-rslora-v7-AWQ",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "tellang/yeji-8b-rslora-v7-AWQ" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tellang/yeji-8b-rslora-v7-AWQ",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use tellang/yeji-8b-rslora-v7-AWQ with Docker Model Runner:
```
docker model run hf.co/tellang/yeji-8b-rslora-v7-AWQ
```

YEJI-8B-RSLoRA-v7-AWQ

yeji-8b-rslora-v7의 AWQ 4-bit 양자화 버전.

Model Description

YEJI 8B 주력 모델의 AWQ 양자화 버전입니다. 원본 8.2B 파라미터를 4-bit로 양자화하여 GPU 메모리 사용량을 대폭 줄이고 vLLM 서빙에 최적화했습니다.

Original Model: tellang/yeji-8b-rslora-v7
Quantization: AWQ 4-bit (group_size=128)
Optimized for: vLLM, TGI

Usage

vLLM (권장)

python -m vllm.entrypoints.openai.api_server \
    --model tellang/yeji-8b-rslora-v7-AWQ \
    --quantization awq \
    --max-model-len 4096

Transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained(
    "tellang/yeji-8b-rslora-v7-AWQ",
    torch_dtype="auto",
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained("tellang/yeji-8b-rslora-v7-AWQ")

All YEJI Models

Model	Params	Type	Downloads
yeji-8b-rslora-v7	8.2B	Full	345
yeji-8b-rslora-v7-AWQ	~2.5B	AWQ 4-bit	371
yeji-4b-instruct-v9	4.0B	Full	65
yeji-4b-instruct-v9-AWQ	~1.3B	AWQ 4-bit	138

Limitations

한국어 운세/점술 도메인 특화. AWQ 양자화로 원본 대비 미세한 품질 저하 가능.
점술 결과는 엔터테인먼트 목적입니다.

Downloads last month: -

Safetensors

Model size

2B params

Tensor type

BF16

I64

I32

Model tree for tellang/yeji-8b-rslora-v7-AWQ

Base model

Qwen/Qwen3-8B-Base

Finetuned

tellang/yeji-8b-rslora-v7

Quantized

(1)

this model

tellang
/

yeji-8b-rslora-v7-AWQ