xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-09 17:05:37 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Neuron] Update Dockerfile.neuron to use latest neuron release (2.23) (#18512)

Browse Source 
Signed-off-by: Satyajith Chilappagari <satchill@amazon.com>
This commit is contained in:
Satyajith Chilappagari 2025-05-22 02:17:34 -07:00 committed by GitHub
parent 23b67b37b2
commit e2d7d31244
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 4 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
docker/Dockerfile.neuron
View File

@ -1,6 +1,6 @@
# default base image
# https://gallery.ecr.aws/neuron/pytorch-inference-neuronx
ARG BASE_IMAGE="public.ecr.aws/neuron/pytorch-inference-neuronx:2.5.1-neuronx-py310-sdk2.22.0-ubuntu22.04"
ARG BASE_IMAGE="public.ecr.aws/neuron/pytorch-inference-neuronx:2.6.0-neuronx-py310-sdk2.23.0-ubuntu22.04"
FROM $BASE_IMAGE
@ -22,8 +22,7 @@ WORKDIR ${APP_MOUNT}/vllm
RUN python3 -m pip install --upgrade pip
RUN python3 -m pip install --no-cache-dir fastapi ninja tokenizers pandas tenacity
RUN python3 -m pip install sentencepiece transformers==4.48.0 -U
RUN python3 -m pip install neuronx-cc==2.17.194.0 --extra-index-url=https://pip.repos.neuron.amazonaws.com -U
RUN python3 -m pip install neuronx-cc==2.* --extra-index-url=https://pip.repos.neuron.amazonaws.com -U
RUN python3 -m pip install pytest
# uninstall transformers-neuronx package explicitly to avoid version conflict
@ -49,6 +48,8 @@ RUN python3 -m pip install -e tests/vllm_test_utils
# FIXME: `--no-deps` argument is temporarily added to resolve transformers package version conflict
RUN python3 -m pip install transformers-neuronx==0.13.* --extra-index-url=https://pip.repos.neuron.amazonaws.com -U --no-deps
RUN python3 -m pip install sentencepiece transformers==4.48.0 -U
# overwrite entrypoint to run bash script
RUN echo "import subprocess; import sys; subprocess.check_call(sys.argv[1:])" > /usr/local/bin/dockerd-entrypoint.py