mirror of
https://github.com/netbirdio/gvisor.git
synced 2026-05-22 17:12:49 -07:00
Otto Bittner
2cf6705467
The benchmark loads a small model to allow running on many platforms and to keep execution time low for experimentation. It will make sense to add a version to this test that loads a larger model (e.g. llama3-8b).
22 lines
1.2 KiB
Docker
22 lines
1.2 KiB
Docker
# v2.45.1
|
|
FROM alpine/git@sha256:4d7fe8d770483993c0cec264d49a573bac49e5239db47a9846572352e72da49c as downloader
|
|
# post checkout hook. checks that command git lfs is available.
|
|
RUN apk add git-lfs && \
|
|
git lfs install && \
|
|
GIT_CLONE_PROTECTION_ACTIVE=false git clone https://huggingface.co/facebook/opt-125m /opt-125m && \
|
|
rm /opt-125m/flax_model.msgpack && \
|
|
rm /opt-125m/tf_model.h5
|
|
# post checkout hook. checks that command git lfs is available.
|
|
RUN GIT_CLONE_PROTECTION_ACTIVE=false git clone https://huggingface.co/datasets/anon8231489123/ShareGPT_Vicuna_unfiltered /dataset
|
|
RUN git clone https://github.com/vllm-project/vllm.git /vllm && cd /vllm && git checkout v0.4.2
|
|
|
|
# v0.4.2
|
|
FROM vllm/vllm-openai@sha256:c63bd9e99c6d5914032b396be38a80fd9640693da62b64b907434d38bf38a8e9
|
|
COPY --from=downloader /vllm/examples/template_chatml.jinja /vllm/examples/template_chatml.jinja
|
|
COPY --from=downloader /vllm/benchmarks /vllm/benchmarks
|
|
COPY --from=downloader /opt-125m /model
|
|
COPY --from=downloader /dataset/ShareGPT_V3_unfiltered_cleaned_split.json /ShareGPT_V3_unfiltered_cleaned_split.json
|
|
|
|
ENTRYPOINT ["python3"]
|
|
CMD ["-m", "vllm.entrypoints.openai.api_server", "--model", "/model", "--chat-template", "/vllm/examples/template_chatml.jinja"]
|