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: move ministral_8b_fp8 to fp8_specific gpu list(exclude Ampere) (#6533)

Browse Source 
Signed-off-by: ruodil <200874449+ruodil@users.noreply.github.com>
Co-authored-by: Larry <197874197+LarryXFly@users.noreply.github.com>
This commit is contained in:
ruodil 2025-08-04 13:22:39 +08:00 committed by GitHub
parent 59d91b8b94
commit 6459725bf9
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 6 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
tests/integration/test_lists/qa/trt_llm_release_perf_test.yml
View File

@ -44,12 +44,6 @@ trt_llm_release_perf_test:
- perf/test_perf.py::test_perf[ministral_8b-bench-pytorch-bfloat16-maxnt:5000-input_output_len:5000,500-reqs:500-con:250]
- perf/test_perf.py::test_perf[ministral_8b-bench-pytorch-bfloat16-input_output_len:500,2000-reqs:500-con:250]
# Ministral-8B FP8
- perf/test_perf.py::test_perf[ministral_8b_fp8-bench-pytorch-float8-maxbs:1-maxnt:5000-input_output_len:5000,500-reqs:8-con:1]
- perf/test_perf.py::test_perf[ministral_8b_fp8-bench-pytorch-float8-maxbs:1-input_output_len:500,2000-reqs:8-con:1]
- perf/test_perf.py::test_perf[ministral_8b_fp8-bench-pytorch-float8-maxnt:5000-input_output_len:5000,500-reqs:500-con:250]
- perf/test_perf.py::test_perf[ministral_8b_fp8-bench-pytorch-float8-input_output_len:500,2000-reqs:500-con:250]
# Ministral-8B LoRA tests (using dummy Mistral LoRA checkpoint)
- perf/test_perf.py::test_perf[ministral_8b-bench-pytorch-bfloat16-maxbs:2-maxnt:1024-input_output_len:128,128-loras:1-reqs:8-con:2]
@ -188,6 +182,12 @@ trt_llm_release_perf_test:
- perf/test_perf.py::test_perf[bielik_11b_v2.2_instruct_fp8-bench-pytorch-float8-input_output_len:1000,1000-con:250]
- perf/test_perf.py::test_perf[bielik_11b_v2.2_instruct_fp8-bench-pytorch-float8-input_output_len:2000,2000-con:250]
# Ministral-8B FP8
- perf/test_perf.py::test_perf[ministral_8b_fp8-bench-pytorch-float8-maxbs:1-maxnt:5000-input_output_len:5000,500-reqs:8-con:1]
- perf/test_perf.py::test_perf[ministral_8b_fp8-bench-pytorch-float8-maxbs:1-input_output_len:500,2000-reqs:8-con:1]
- perf/test_perf.py::test_perf[ministral_8b_fp8-bench-pytorch-float8-maxnt:5000-input_output_len:5000,500-reqs:500-con:250]
- perf/test_perf.py::test_perf[ministral_8b_fp8-bench-pytorch-float8-input_output_len:500,2000-reqs:500-con:250]
- condition:
terms:
supports_fp8: true