-
Notifications
You must be signed in to change notification settings - Fork 1k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
add minicpm3 4b #633
add minicpm3 4b #633
Conversation
add minicpm3 model
add minicpm3 4b
minicpm3-4b-result.zip |
Apologize for the long delay. Will definitely take a look after the Monday ICLR deadline. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hi @Cppowboy,
Thanks for the PR.
Instead of cloning the MiniCPM repo and spinning up the vllm server through your script, I follow your chat template on huggingface and add the MiniCPM handler
for local vllm inference (which already has all the vllm server stuff handled).
add minicpm3 4b model # start a minicpm3 4b function call server ```shell git clone https://github.com/OpenBMB/MiniCPM.git cd MiniCPM/demo/minicpm3/function_call pip install vllm==0.5.4 pip install datamodel_code_generator python openai_api_server.py --model openbmb/MiniCPM3-4B --chat-template chatml.jinja --dtype auto --api-key token-abc123 --tensor-parallel-size 1 --trust-remote-code ``` # run evaluation ```shell export OPENAI_API_KEY=token-abc123 export OPENAI_BASE_URL=http://127.0.0.1:8000/v1 python openfunctions_evaluation.py --model openbmb/MiniCPM3-4B-FC --test-category all ``` --------- Co-authored-by: Huanzhi (Hans) Mao <[email protected]>
add minicpm3 4b model
start a minicpm3 4b function call server
git clone https://github.com/OpenBMB/MiniCPM.git cd MiniCPM/demo/minicpm3/function_call pip install vllm==0.5.4 pip install datamodel_code_generator python openai_api_server.py --model openbmb/MiniCPM3-4B --chat-template chatml.jinja --dtype auto --api-key token-abc123 --tensor-parallel-size 1 --trust-remote-code
run evaluation