@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / a985f7af9f
Fork 0

A high-throughput and memory-efficient inference and serving engine for LLMs

Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git

Synced Nov 16, 04:30 UTC

[CI] Adding CPU docker pipeline (#11261)

Signed-off-by: Yuan Zhou <yuan.zhou@intel.com> Co-authored-by: Kevin H. Luu <kevin@anyscale.com>

Yuan committed Dec 20, 2024 at 03:46 UTC a985f7af9f7b249974b283a9d999575ac30fac3d
1 file changed +15
.buildkite/release-pipeline.yaml
+15
@@ -55,3 +55,18 @@ steps:
55 password-env: DOCKERHUB_TOKEN
56 env:
57 DOCKER_BUILDKIT: "1"
58 +
59 + - block: "Build CPU release image"
60 + key: block-cpu-release-image-build
61 + depends_on: ~
62 +
63 + - label: "Build and publish CPU release image"
64 + depends_on: block-cpu-release-image-build
65 + agents:
66 + queue: cpu_queue_postmerge
67 + commands:
68 + - "aws ecr-public get-login-password --region us-east-1 | docker login --username AWS --password-stdin public.ecr.aws/q9t5s3a7"
69 + - "DOCKER_BUILDKIT=1 docker build --build-arg max_jobs=16 --build-arg GIT_REPO_CHECK=1 --tag public.ecr.aws/q9t5s3a7/vllm-cpu-release-repo:$RELEASE_VERSION --progress plain -f Dockerfile.cpu ."
70 + - "docker push public.ecr.aws/q9t5s3a7/vllm-cpu-release-repo:$RELEASE_VERSION"
71 + env:
72 + DOCKER_BUILDKIT: "1"