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 5, 03:30 UTC
Bug/test eagle dp v2 (#39838)
Signed-off-by: Monishver Chandrasekaran <monishverchandrasekaran@gmail.com>
Monishver committed
Apr 15, 2026 at 13:48 UTC
21e5a9f48e773e36e916bc8d10c4ab4aed3887a7
1 file changed
-1
.buildkite/test_areas/distributed.yaml
-1
@@ -196,7 +196,6 @@ steps:
196
- VLLM_ALLOW_INSECURE_SERIALIZATION=1 python3 examples/rl/rlhf_async_new_apis.py
197
- 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
198
- pytest -v -s tests/v1/distributed/test_dbo.py
199
- - TP_SIZE=1 DP_SIZE=2 pytest -v -s tests/v1/distributed/test_eagle_dp.py
199
200
- label: Distributed Tests (2 GPUs)(B200)
201
device: b200