Instructions to use coyude/Chinese-Pygmalion-13b-GPTQ with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use coyude/Chinese-Pygmalion-13b-GPTQ with Transformers:

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

pipe = pipeline("text-generation", model="coyude/Chinese-Pygmalion-13b-GPTQ")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("coyude/Chinese-Pygmalion-13b-GPTQ")
model = AutoModelForCausalLM.from_pretrained("coyude/Chinese-Pygmalion-13b-GPTQ")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use coyude/Chinese-Pygmalion-13b-GPTQ with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "coyude/Chinese-Pygmalion-13b-GPTQ"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "coyude/Chinese-Pygmalion-13b-GPTQ",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/coyude/Chinese-Pygmalion-13b-GPTQ

SGLang

How to use coyude/Chinese-Pygmalion-13b-GPTQ 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 "coyude/Chinese-Pygmalion-13b-GPTQ" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "coyude/Chinese-Pygmalion-13b-GPTQ",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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 "coyude/Chinese-Pygmalion-13b-GPTQ" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "coyude/Chinese-Pygmalion-13b-GPTQ",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use coyude/Chinese-Pygmalion-13b-GPTQ with Docker Model Runner:
```
docker model run hf.co/coyude/Chinese-Pygmalion-13b-GPTQ
```

原始模型：https://huggingface.co/TehVenom/Pygmalion-13b-Merged

lora：https://huggingface.co/ziqingyang/chinese-alpaca-lora-13b

将Pygmalion-13b与chinese-alpaca-lora-13b进行合并，增强模型的中文能力，~~不过存在翻译腔~~

使用项目： https://github.com/ymcui/Chinese-LLaMA-Alpaca

https://github.com/qwopqwop200/GPTQ-for-LLaMa

兼容AutoGPTQ和GPTQ-for-LLaMa
若选择GPTQ-for-LLaMa加载，请设置 Wbits=4 groupsize=128 model_type=llama

Text-generation-webui懒人包： https://www.bilibili.com/read/cv23495183

Original model: https://huggingface.co/TehVenom/Pygmalion-13b-Merged

lora：https://huggingface.co/ziqingyang/chinese-alpaca-lora-13b

The Pygmalion-13b model is combined with the chinese-alpaca-lora-13b model to enhance the model's Chinese language capabilities, ~~although there may be some translated tone~~.

Usage projects: https://github.com/ymcui/Chinese-LLaMA-Alpaca

https://github.com/qwopqwop200/GPTQ-for-LLaMa

Compatible with AutoGPTQ and GPTQ-for-LLaMa
If you choose to load GPTQ-for-LLaMa, please set Wbits=4 groupsize=128 model_type=llama

Downloads last month: 8