Instructions to use charent/ChatLM-mini-Chinese with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use charent/ChatLM-mini-Chinese with Transformers:

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

pipe = pipeline("text-generation", model="charent/ChatLM-mini-Chinese", trust_remote_code=True)

# Load model directly
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

tokenizer = AutoTokenizer.from_pretrained("charent/ChatLM-mini-Chinese", trust_remote_code=True)
model = AutoModelForSeq2SeqLM.from_pretrained("charent/ChatLM-mini-Chinese", trust_remote_code=True)

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use charent/ChatLM-mini-Chinese with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/charent/ChatLM-mini-Chinese

SGLang

How to use charent/ChatLM-mini-Chinese 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 "charent/ChatLM-mini-Chinese" \
    --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": "charent/ChatLM-mini-Chinese",
		"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 "charent/ChatLM-mini-Chinese" \
        --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": "charent/ChatLM-mini-Chinese",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use charent/ChatLM-mini-Chinese with Docker Model Runner:
```
docker model run hf.co/charent/ChatLM-mini-Chinese
```

Thank you very much for your excellent work!非常感谢您的出色的工作！

by wukongai - opened Jan 20, 2024

Discussion

wukongai

Jan 20, 2024

This can serve as a starter learning model!
这个可以作为一个入门的学习模型！

wukongai

Jan 20, 2024

另外，能否提供一下q8、q16的gguf量化权重，我想在cpu上跑一下，肯定非常欢快！

Cran-May

Jan 21, 2024

另外，能否提供一下q8、q16的gguf量化权重，我想在cpu上跑一下，肯定非常欢快！

But t5 models seem not to be able to transfer to GGUF format.
t5模型应该不能gguf量化，至少llama.cpp对其没有支持

charent

Owner Jan 21, 2024

•

edited Jan 22, 2024

另外，能否提供一下q8、q16的gguf量化权重，我想在cpu上跑一下，肯定非常欢快！

暂时没有发布量化模型的计划，发布的是float32权重，可以直接在cpu上进行推理，200M的参数在cpu上推理已经很快了。如果有需要，你可以自己做量化（可能需要cuda设备），安装'bitsandbytes':

# linux:  
pip install bitsandbytes

# windows
python -m pip install bitsandbytes --prefer-binary --extra-index-url=https://jllllll.github.io/bitsandbytes-windows-webui

8bit量化代码如下：

from transformers import AutoModelForSeq2SeqLM
model = AutoModelForSeq2SeqLM.from_pretrained('charent/ChatLM-mini-Chinese', trust_remote_code=True, device_map="auto",  load_in_8bit=True)
model.save_pretrained('./model_8bit')

wukongai

Jan 22, 2024

收到，谢谢！

zhaochaofeng

May 18, 2024

请问一下，RLHF训练过程的奖励模型在哪里体现呢？

FiveAirplane68

Jun 21, 2024

好兄弟这实在是泰裤辣❤❤❤！！！

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment