kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-01-14 06:27:45 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

test: add deepseek_v3_lite rcca cases (#5225)

Browse Source 
Signed-off-by: ruodil <200874449+ruodil@users.noreply.github.com>
This commit is contained in:
ruodil 2025-06-16 13:39:26 +08:00 committed by GitHub
parent cbc6455266
commit e05b3ff427
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
tests/integration/defs/perf/model_yaml_config.py
View File

@ -78,7 +78,7 @@ def get_model_yaml_config(model_label: str) -> dict:
[1, 2, 4, 8, 16, 32, 64, 128, 256, 384] [1, 2, 4, 8, 16, 32, 64, 128, 256, 384]
}, },
}, },
'deepseek_r1_nvfp4-bench-pytorch-streaming-float4-maxbs:2048-maxnt:8192-input_output_len:256,256-num_reqs:200-gpus:1': 'deepseek_v3_lite_nvfp4-bench-pytorch-streaming-float4-maxbs:2048-maxnt:8192-input_output_len:256,256-reqs:200':
{ {
'pytorch_backend_config': { 'pytorch_backend_config': {
'print_iter_log': True, 'print_iter_log': True,

2
tests/integration/test_lists/qa/trt_llm_release_perf_cluster_test.yml
View File

@ -11,7 +11,7 @@ trt_llm_release_perf_cluster_test:
- perf/test_perf.py::test_perf[t5-bench-float16-input_output_len:128,20] - perf/test_perf.py::test_perf[t5-bench-float16-input_output_len:128,20]
- perf/test_perf.py::test_perf[mistral_7b_v0.1-bench-float16-input_output_len:1000,1000-quant:fp8] - perf/test_perf.py::test_perf[mistral_7b_v0.1-bench-float16-input_output_len:1000,1000-quant:fp8]
- perf/test_perf.py::test_perf[mistral_7b_v0.1-bench-float16-input_output_len:500,2000-quant:fp8] - perf/test_perf.py::test_perf[mistral_7b_v0.1-bench-float16-input_output_len:500,2000-quant:fp8]
- perf/test_perf.py::test_perf[deepseek_r1_nvfp4-bench-pytorch-streaming-float4-maxbs:2048-maxnt:8192-input_output_len:256,256-num_reqs:200-gpus:1] - perf/test_perf.py::test_perf[deepseek_v3_lite_nvfp4-bench-pytorch-streaming-float4-maxbs:2048-maxnt:8192-input_output_len:256,256-reqs:200-gpus:1]
- condition: - condition:
ranges: ranges:
system_gpu_count: system_gpu_count: