kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-02-18 16:55:08 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

tests: update api change from decoder to sampler in test (#4479)

Browse Source 
update

Signed-off-by: Ivy Zhang <25222398+crazydemo@users.noreply.github.com>
This commit is contained in:
Ivy Zhang 2025-05-21 14:22:18 +08:00 committed by GitHub
parent cc3f8e6431
commit e977c75300
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 4 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

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

@ -187,10 +187,11 @@ class TestLlama3_1_8BInstruct(LlmapiAccuracyTestHarness):
task = GSM8K(self.MODEL_NAME)
task.evaluate(llm)
@pytest.mark.skip(reason="https://nvbugspro.nvidia.com/bug/5292517")
@skip_pre_hopper
def test_fp8_llm_decoder(self):
def test_fp8_llm_sampler(self):
model_path = f"{llm_models_root()}/llama-3.1-model/Llama-3.1-8B-Instruct-FP8"
pytorch_config = PyTorchConfig(enable_trtllm_decoder=True)
pytorch_config = PyTorchConfig(enable_trtllm_sampler=True)
llm = LLM(model_path, pytorch_backend_config=pytorch_config)
assert llm.args.quant_config.quant_algo == QuantAlgo.FP8

2
tests/integration/test_lists/qa/examples_test_list.txt
View File

@ -425,7 +425,7 @@ accuracy/test_llm_api.py::TestMixtral8x7B::test_tp2
accuracy/test_llm_api.py::TestMixtral8x7B::test_smooth_quant_tp2pp2
accuracy/test_llm_api.py::TestMixtral8x7BInstruct::test_awq_tp2
accuracy/test_llm_api_pytorch.py::TestLlama3_1_8B::test_nvfp4
accuracy/test_llm_api_pytorch.py::TestLlama3_1_8BInstruct::test_fp8_llm_decoder
accuracy/test_llm_api_pytorch.py::TestLlama3_1_8BInstruct::test_fp8_llm_sampler
accuracy/test_llm_api_pytorch.py::TestLlama3_3_70BInstruct::test_fp8_tp4
accuracy/test_llm_api_pytorch.py::TestLlama3_3_70BInstruct::test_nvfp4_tp4
accuracy/test_cli_flow.py::TestLlama3_3_70BInstruct::test_fp8_prequantized_tp4