Instructions to use mymusise/CPM-Generate-distill with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mymusise/CPM-Generate-distill with Transformers:

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

pipe = pipeline("text-generation", model="mymusise/CPM-Generate-distill")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mymusise/CPM-Generate-distill")
model = AutoModelForCausalLM.from_pretrained("mymusise/CPM-Generate-distill")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use mymusise/CPM-Generate-distill with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/mymusise/CPM-Generate-distill

SGLang

How to use mymusise/CPM-Generate-distill 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 "mymusise/CPM-Generate-distill" \
    --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": "mymusise/CPM-Generate-distill",
		"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 "mymusise/CPM-Generate-distill" \
        --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": "mymusise/CPM-Generate-distill",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use mymusise/CPM-Generate-distill with Docker Model Runner:
```
docker model run hf.co/mymusise/CPM-Generate-distill
```

CPM-Generate-distill

CPM(Chinese Pre-Trained Language Models), which has 2.6B parameters, made by the research team of Beijing Zhiyuan Institute of artificial intelligence and Tsinghua University @TsinghuaAI.

repo: CPM-Generate The One Thing You Need to Know is this model is not uploaded by official, the conver script is here

And the CPM-Generate-distill is the distill model of CPM.

How to use

How to use this model directly from the 🤗/transformers library:

from transformers import TextGenerationPipeline, AutoTokenizer, AutoModelWithLMHead

tokenizer = AutoTokenizer.from_pretrained("mymusise/CPM-Generate-distill")
model = AutoModelWithLMHead.from_pretrained("mymusise/CPM-Generate-distill")

text_generater = TextGenerationPipeline(model, tokenizer)

print(text_generator('清华大学是', max_length=50, do_sample=True, top_p=0.9))

Downloads last month: 11