[TPU] Support tensor parallelism in async llm engine (#6891)

This commit is contained in:
Earthwalker
2024-07-29 12:42:21 -07:00
committed by GitHub
parent 60d1c6e584
commit 7f8d612d24
2 changed files with 11 additions and 2 deletions
+3
View File
@@ -12,6 +12,9 @@ RUN pip install "numpy<2"
RUN pip install torch_xla[tpu] -f https://storage.googleapis.com/libtpu-releases/index.html
RUN pip install torch_xla[pallas] -f https://storage.googleapis.com/jax-releases/jax_nightly_releases.html -f https://storage.googleapis.com/jax-releases/jaxlib_nightly_releases.html
# Fix FastAPI dependence
RUN pip install "starlette<0.38.0"
# Build vLLM.
COPY . /workspace/vllm
ENV VLLM_TARGET_DEVICE="tpu"