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 14, 05:30 UTC
chore: build release image by default (#23852)
Signed-off-by: Codex <codex@openai.com>
Simon Mo committed
Aug 28, 2025 at 13:17 UTC
27e88cee748d41e07268ca140d15252c6b38acf1
1 file changed
+2
-6
.buildkite/release-pipeline.yaml
+2
-6
@@ -62,12 +62,8 @@ steps:
62
env:
63
DOCKER_BUILDKIT: "1"
64
65
- - block: "Build release image (x86)"
66
- depends_on: ~
67
- key: block-release-image-build
68
-
65
- label: "Build release image (x86)"
70
- depends_on: block-release-image-build
66
+ depends_on: ~
67
id: build-release-image-x86
68
agents:
69
queue: cpu_queue_postmerge
@@ -80,7 +76,7 @@ steps:
76
- "docker push public.ecr.aws/q9t5s3a7/vllm-release-repo:$BUILDKITE_COMMIT"
77
78
- label: "Build release image (arm64)"
83
- depends_on: block-release-image-build
79
+ depends_on: ~
80
id: build-release-image-arm64
81
agents:
82
queue: arm64_cpu_queue_postmerge