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

[None][test] Remove most TRT-backend test cases in llm_perf_nim.yml (#10572)

Browse Source 
Signed-off-by: Fanrong Li <23290157+lfr-0531@users.noreply.github.com>
Signed-off-by: yufeiwu-nv <230315618+yufeiwu-nv@users.noreply.github.com>
Co-authored-by: Fanrong Li <23290157+lfr-0531@users.noreply.github.com>
This commit is contained in:
yufeiwu-nv 2026-01-12 15:34:55 +08:00 committed by GitHub
parent c5914f9085
commit 8e806abac3
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 1 additions and 17 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
tests/integration/test_lists/qa/llm_perf_sanity.yml
View File

@ -9,7 +9,6 @@ llm_perf_sanity:
# 4: A100, L40S, H20, H100, H200, B200, B300, GB200, GB300, RTX6000-Server
# 5: L40S, H20, H100, H200, B200, B300, RTX6000-Server
# 6: H20, H100, H200, B200, B300, RTX6000-Server
# 7: H20, H100, H200, B200, B300
# ===============================================================================
# 1: All GPUs
@ -31,6 +30,7 @@ llm_perf_sanity:
- perf/test_perf.py::test_perf[llama_v3.1_8b-bench-pytorch-streaming-bfloat16-input_output_len:500,2000]
- perf/test_perf.py::test_perf[llama_v3.1_8b-bench-pytorch-streaming-bfloat16-input_output_len:512,32]
- perf/test_perf.py::test_perf[llama_v3.1_8b-bench-pytorch-bfloat16-input_output_len:128,128]
- perf/test_perf.py::test_perf[starcoder2_7b-bench-pytorch-bfloat16-input_output_len:512,512]
# Phi-4-multimodal-instruct
- perf/test_perf.py::test_perf[phi_4_multimodal_instruct-bench-pytorch-bfloat16-input_output_len:128,128]
# Bielik-11B-v2.2-Instruct
@ -124,25 +124,9 @@ llm_perf_sanity:
# for chunked prefill cases
- perf/test_perf.py::test_perf[deepseek_v3_lite_fp8-bench-pytorch-float8-maxbs:512-maxnt:2048-kv_frac:0.85-input_output_len:3000,500-reqs:200]
- perf/test_perf.py::test_perf[llama_v3.1_405b_instruct_fp8-bench-pytorch-float8-input_output_len:128,128-tp:8-gpus:8]
- perf/test_perf.py::test_perf[llama_v4_maverick_17b_128e_instruct_fp8-bench-pytorch-float8-input_output_len:512,32-ep:8-tp:8-gpus:8] TIMEOUT(100)
- perf/test_perf.py::test_perf[llama_v4_scout_17b_16e_instruct_fp8-bench-pytorch-float8-maxbs:1024-maxnt:20000-kv_frac:0.85-input_output_len:20000,2000-reqs:1000-ep:8-tp:8-gpus:8] TIMEOUT(100)
- perf/test_perf.py::test_perf[qwen3_235b_a22b_fp8-bench-pytorch-float8-input_output_len:1000,2000-con:256-ep:8-gpus:8] TIMEOUT(60)
# disagg server cases
- perf/test_perf.py::test_perf[deepseek_v3_lite_fp8-disagg_server-ctx_dp:4-gen_tp:4]
- perf/test_perf.py::test_perf[llama_v3.1_8b-disagg_server-ctx_dp:4-gen_tp:4]
# gpt_oss_20b_fp4
- perf/test_perf.py::test_perf[gpt_oss_20b_fp4-bench-pytorch-float4-input_output_len:512,512]
# 7: H20, H100, H200, B200, B300
- condition:
ranges:
system_gpu_count:
gte: 8
compute_capability:
gte: 9.0
lt: 12.0
tests:
# chunked attention case
- perf/test_perf.py::test_perf[llama_v4_maverick_17b_128e_instruct_fp8-bench-pytorch-float8-maxbs:1024-maxnt:20000-kv_frac:0.6-input_output_len:20000,2000-reqs:1000-ep:8-tp:8-gpus:8]