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, 00:00 UTC
[ROCm][CI] Fix "Distributed Tests (H200)" Test (#31227)
Signed-off-by: kliuae <kuanfu.liu@embeddedllm.com>
kliuae committed
Dec 24, 2025 at 14:56 UTC
1c74150bcabd1bd4ae36e562aefb73e53e016dba
1 file changed
+1
-1
.buildkite/test-amd.yaml
+1
-1
@@ -1518,7 +1518,7 @@ steps:
1518
- "VLLM_TEST_CLEAN_GPU_MEMORY=1 pytest -v -s tests/compile/distributed/test_fusions_e2e.py -k 'not Llama-4'"
1519
- VLLM_TEST_CLEAN_GPU_MEMORY=1 pytest -v -s tests/distributed/test_sequence_parallel.py
1520
- pytest -v -s tests/distributed/test_context_parallel.py
1521
- - HIP_VISIBLE_DEVICES=0,1 VLLM_USE_DEEP_GEMM=1 VLLM_LOGGING_LEVEL=DEBUG python3 examples/offline_inference/data_parallel.py --model=Qwen/Qwen1.5-MoE-A2.7B -tp=1 -dp=2 --max-model-len=2048 --all2all-backend=deepep_high_throughput
1521
+ - HIP_VISIBLE_DEVICES=0,1 VLLM_LOGGING_LEVEL=DEBUG python3 examples/offline_inference/data_parallel.py --model=Qwen/Qwen1.5-MoE-A2.7B -tp=1 -dp=2 --max-model-len=2048 --all2all-backend=allgather_reducescatter --disable-nccl-for-dp-synchronization
1522
- pytest -v -s tests/v1/distributed/test_dbo.py
1523
1524
##### B200 test #####