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 15, 02:15 UTC
[Test] Remove docker build from test. (#20542)
Signed-off-by: Qiliang Cui <derrhein@gmail.com>
QiliangCui committed
Jul 10, 2025 at 11:21 UTC
c66e38ea4c12651baacd0b9021ef720a5d93a995
1 file changed
-10
.buildkite/scripts/tpu/docker_run_bm.sh
-10
@@ -22,16 +22,6 @@ trap remove_docker_container EXIT
22
# Remove the container that might not be cleaned up in the previous run.
23
remove_docker_container
24
25
-# Build docker image.
26
-# TODO: build the image outside the script and share the image with other
27
-# tpu test if building time is too long.
28
-DOCKER_BUILDKIT=1 docker build \
29
- --build-arg max_jobs=16 \
30
- --build-arg USE_SCCACHE=1 \
31
- --build-arg GIT_REPO_CHECK=0 \
32
- --tag vllm/vllm-tpu-bm \
33
- --progress plain -f docker/Dockerfile.tpu .
34
-
25
LOG_ROOT=$(mktemp -d)
26
# If mktemp fails, set -e will cause the script to exit.
27
echo "Results will be stored in: $LOG_ROOT"