kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-01-13 22:18:36 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

[https://nvbugs/5745152][fix] Fix some GPTOSS test setups (#10085)

Browse Source 
Signed-off-by: Dongfeng Yu <dongfengy@nvidia.com>
This commit is contained in:
dongfengy 2025-12-26 17:52:40 +08:00 committed by GitHub
parent 4a5ef84dc2
commit bfc591994c
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
3 changed files with 7 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
tests/integration/defs/accuracy/test_disaggregated_serving.py
View File

@ -1091,11 +1091,13 @@ class TestGPTOSS(LlmapiAccuracyTestHarness):
"max_attention_window": [128, 32768],
"enable_block_reuse": block_reuse,
"enable_partial_reuse": False,
"free_gpu_memory_fraction": 0.5,
}
gen_server_config["kv_cache_config"] = {
"max_attention_window": [128, 32768],
"enable_block_reuse": block_reuse,
"enable_partial_reuse": False,
"free_gpu_memory_fraction": 0.5,
}
disaggregated_server_config = {
"hostname": "localhost",

5
tests/integration/defs/accuracy/test_llm_api_pytorch.py
View File

@ -4369,6 +4369,11 @@ class TestGPTOSS(LlmapiAccuracyTestHarness):
"https://nvbugs/5636916: Remaining Hopper Eagle Accuracy Issue for only TP=4"
)
if not one_model and overlap_scheduler:
pytest.skip(
"https://nvbugs/5745152: two_model + overlap_scheduler can sometimes time out."
)
MAX_OUTPUT_LEN = 128179
MAX_INPUT_LEN = 32768

2
tests/integration/test_lists/waives.txt
View File

@ -320,8 +320,6 @@ accuracy/test_llm_api_pytorch.py::TestLlama3_1NemotronNano8Bv1::test_fp8_prequan
accuracy/test_llm_api_pytorch.py::TestNemotronH_47B_Base::test_auto_dtype[tp8ep4-cuda_graph=True] SKIP (https://nvbugs/5640697)
accuracy/test_llm_api_pytorch.py::TestNemotronH_47B_Base::test_reasoning_fp8_prequantized[tp8ep8-cuda_graph=True] SKIP (https://nvbugs/5640697)
accuracy/test_llm_api_pytorch.py::TestQwQ_32B::test_auto_dtype_tp4 SKIP (https://nvbugs/5640697)
accuracy/test_disaggregated_serving.py::TestGPTOSS::test_auto_dtype[True] SKIP (https://nvbugs/5644632)
accuracy/test_disaggregated_serving.py::TestGPTOSS::test_auto_dtype[False] SKIP (https://nvbugs/5644632)
test_e2e.py::test_ptp_quickstart_multimodal[mistral-small-3.1-24b-instruct-Mistral-Small-3.1-24B-Instruct-2503-image-True] SKIP (https://nvbugs/5648560)
test_e2e.py::test_ptp_quickstart_multimodal[mistral-small-3.1-24b-instruct-Mistral-Small-3.1-24B-Instruct-2503-image-False] SKIP (https://nvbugs/5648560)
accuracy/test_llm_api_pytorch.py::TestDeepSeekR1::test_nvfp4_multi_gpus[latency_trtllmgen_adp_lmtp] SKIP (https://nvbugs/5629136)