kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-02-05 02:31:33 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

[None][test] Update test case for release (#10763)

Browse Source 
Signed-off-by: Ivy Zhang <25222398+crazydemo@users.noreply.github.com>
Signed-off-by: Wangshanshan <30051912+dominicshanshan@users.noreply.github.com>
This commit is contained in:
Ivy Zhang 2026-01-19 16:36:07 +08:00 committed by Yanchao Lu
parent 4df0ca8bd1
commit 4ebc1b1596
2 changed files with 3 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

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

@ -452,7 +452,7 @@ class LongBenchV2(AccuracyTask):
EVALUATOR_KWARGS = dict(
dataset_path=DATASET_DIR,
length="medium",
max_len=120000,
max_input_length=120000,
apply_chat_template=True,
random_seed=0,
)

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

@ -1126,6 +1126,7 @@ class TestGemma3_1BInstruct(LlmapiAccuracyTestHarness):
task = MMLU(self.MODEL_NAME)
task.evaluate(llm)
@skip_pre_hopper
def test_fp8_vswa_reuse(self):
# NOTE: Test with VSWA kv cache config.
kv_cache_config = KvCacheConfig(
@ -1140,6 +1141,7 @@ class TestGemma3_1BInstruct(LlmapiAccuracyTestHarness):
task = MMLU(self.MODEL_NAME)
task.evaluate(llm)
@skip_pre_hopper
@pytest.mark.parametrize("backend", ["xgrammar"])
def test_fp8_guided_decoding_vswa_reuse(self, backend: str, mocker):
mocker.patch.dict(os.environ, {"TRTLLM_XGUIDANCE_LENIENT": "1"})