Instructions to use internlm/internlm-20b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use internlm/internlm-20b with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="internlm/internlm-20b", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("internlm/internlm-20b", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use internlm/internlm-20b with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "internlm/internlm-20b" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "internlm/internlm-20b", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/internlm/internlm-20b
- SGLang
How to use internlm/internlm-20b 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 "internlm/internlm-20b" \ --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": "internlm/internlm-20b", "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 "internlm/internlm-20b" \ --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": "internlm/internlm-20b", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use internlm/internlm-20b with Docker Model Runner:
docker model run hf.co/internlm/internlm-20b
commercial license
I used the form to request a commercial license like a week ago, was wondering about what time to wait to expect.
I used the form to request a commercial license like a week ago, was wondering about what time to wait to expect.
Thank you for your attention. Due to the National Day holiday in China from September 29th to October 6th, the progress has been slowed down. We will expedite the process once we resume work. Please kindly wait for nearly one more week, and we will get back to you via email.
Would be great if the community could use the infrastructure around the llama models for internlm, too!
https://huggingface.co/internlm/internlm-7b/discussions/4/files
I think that would push adoption even more!
You could consider releasing a second, llamafied version :)
It's not just about the architecture or which infrastructure is best, it's also about being opportunistic about the attentional ecosystems, using the "buzz-infrastructure" for your own benefit.
