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 12, 17:15 UTC
[CI/Build] Disable test_gptoss_tp.py in 'LoRA TP Test' group for ROCm platform (#29204)
Signed-off-by: qli88 <qiang.li2@amd.com>
qli88 committed
Nov 21, 2025 at 19:36 UTC
1ef9c9e29480f95340e124cc7d81a2876a60516d
1 file changed
+4
-1
.buildkite/test-amd.yaml
+4
-1
@@ -1319,7 +1319,10 @@ steps:
1319
- pytest -v -s -x lora/test_llama_tp.py
1320
- pytest -v -s -x lora/test_llm_with_multi_loras.py
1321
- pytest -v -s -x lora/test_olmoe_tp.py
1322
- - pytest -v -s -x lora/test_gptoss_tp.py
1322
+
1323
+ # Disabled for now because MXFP4 backend on non-cuda platform
1324
+ # doesn't support LoRA yet
1325
+ #- pytest -v -s -x lora/test_gptoss_tp.py
1326
1327
1328
- label: Weight Loading Multiple GPU Test # 33min