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 11, 01:15 UTC
[AMD][CI] Add "V1 Test e2e + engine" to mi325_8 Agent Pool (#31040)
Signed-off-by: Micah Williamson <micah.williamson@amd.com>
Micah Williamson committed
Dec 22, 2025 at 09:41 UTC
a5bc77c253a6a3a7f2df6936ac7c9f233030df09
1 file changed
+3
-1
.buildkite/test-amd.yaml
+3
-1
@@ -349,7 +349,9 @@ steps:
349
- label: V1 Test e2e + engine # 65min
350
timeout_in_minutes: 90
351
mirror_hardwares: [amdexperimental]
352
- agent_pool: mi325_4
352
+ # The test uses 4 GPUs, but we schedule it on 8-GPU machines for stability.
353
+ # See discussion here: https://github.com/vllm-project/vllm/pull/31040
354
+ agent_pool: mi325_8
355
# grade: Blocking
356
source_file_dependencies:
357
- vllm/