[TRTLLM-5252][test] add for mistral_small_3.1_24b perf test (#6685)

Signed-off-by: ruodil <[email protected]>

Author: ruodil

tests/integration/defs/perf/test_perf.py

@@ -127,6 +127,7 @@
     "phi_4_multimodal_instruct_audio": "multimodals/Phi-4-multimodal-instruct",
     "bielik_11b_v2.2_instruct": "Bielik-11B-v2.2-Instruct",
     "bielik_11b_v2.2_instruct_fp8": "Bielik-11B-v2.2-Instruct-FP8",
+    "mistral_small_v3.1_24b": "Mistral-Small-3.1-24B-Instruct-2503",
 }
 # Model PATH of HuggingFace
 HF_MODEL_PATH = {

tests/integration/test_lists/qa/llm_perf_cluster.yml

@@ -42,6 +42,11 @@ llm_perf_cluster:
   - 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]
+  #Mistral-Small-3.1-24B-Instruct-2503
+  - perf/test_perf.py::test_perf[mistral_small_v3.1_24b-bench-pytorch-bfloat16-maxbs:1-input_output_len:1000,2000-reqs:8-con:1]
+  - perf/test_perf.py::test_perf[mistral_small_v3.1_24b-bench-pytorch-bfloat16-input_output_len:1000,2000-reqs:500-con:200]
+  - perf/test_perf.py::test_perf[mistral_small_v3.1_24b-bench-pytorch-bfloat16-maxbs:1-maxnt:20000-input_output_len:20000,2000-reqs:8-con:1] TIMEOUT(120)
+  - perf/test_perf.py::test_perf[mistral_small_v3.1_24b-bench-pytorch-bfloat16-maxbs:4096-maxnt:20000-input_output_len:20000,2000-reqs:500-con:200] TIMEOUT(120)
 
 
 - condition:
@@ -57,6 +62,11 @@ llm_perf_cluster:
   - perf/test_perf.py::test_perf[llama_v3.3_70b_instruct_fp8-bench-pytorch-streaming-float8-maxbs:256-input_output_len:512,32-gpus:2]
   - perf/test_perf.py::test_perf[llama_v2_13b-bench-float16-input_output_len:128,128-loras:8-gpus:2]
   - perf/test_perf.py::test_perf[mixtral_8x7b_v0.1-bench-float16-input_output_len:128,128-quant:fp8-gpus:2]
+  #Mistral-Small-3.1-24B-Instruct-2503
+  - perf/test_perf.py::test_perf[mistral_small_v3.1_24b-bench-pytorch-bfloat16-maxbs:1-input_output_len:1000,2000-reqs:8-con:1-gpus:2]
+  - perf/test_perf.py::test_perf[mistral_small_v3.1_24b-bench-pytorch-bfloat16-input_output_len:1000,2000-reqs:500-con:200-gpus:2]
+  - perf/test_perf.py::test_perf[mistral_small_v3.1_24b-bench-pytorch-bfloat16-maxbs:1-maxnt:20000-input_output_len:20000,2000-reqs:8-con:1-gpus:2] TIMEOUT(120)
+  - perf/test_perf.py::test_perf[mistral_small_v3.1_24b-bench-pytorch-bfloat16-maxbs:4096-maxnt:20000-input_output_len:20000,2000-reqs:500-con:200-gpus:2] TIMEOUT(120)
 
 # Tests for systems with 4+ GPUs
 - condition:

tests/integration/test_lists/qa/llm_perf_full.yml

@@ -49,6 +49,11 @@ llm_perf_full:
   - perf/test_perf.py::test_perf[phi_4_mini_instruct-bench-pytorch-bfloat16-input_output_len:2000,500]
   - perf/test_perf.py::test_perf[phi_4_mini_instruct-bench-pytorch-bfloat16-input_output_len:128,128]
   - perf/test_perf.py::test_perf[phi_4_mini_instruct-bench-pytorch-bfloat16-input_output_len:512,32]
+  #Mistral-Small-3.1-24B-Instruct-2503
+  - perf/test_perf.py::test_perf[mistral_small_v3.1_24b-bench-pytorch-bfloat16-maxbs:1-input_output_len:1000,2000-reqs:8-con:1]
+  - perf/test_perf.py::test_perf[mistral_small_v3.1_24b-bench-pytorch-bfloat16-input_output_len:1000,2000-reqs:500-con:200]
+  - perf/test_perf.py::test_perf[mistral_small_v3.1_24b-bench-pytorch-bfloat16-maxbs:1-maxnt:20000-input_output_len:20000,2000-reqs:8-con:1] TIMEOUT(120)
+  - perf/test_perf.py::test_perf[mistral_small_v3.1_24b-bench-pytorch-bfloat16-maxbs:4096-maxnt:20000-input_output_len:20000,2000-reqs:500-con:200] TIMEOUT(120)
 
   # Test list validation
   - test_list_validation.py::test_list_validation
@@ -131,6 +136,11 @@ llm_perf_full:
   - perf/test_perf.py::test_perf[llama_v3.2_1b-bench-pytorch-bfloat16-input_output_len:512,32-gpus:2]
   - perf/test_perf.py::test_perf[llama_v3.2_1b-bench-pytorch-bfloat16-input_output_len:512,200-gpus:2]
   - perf/test_perf.py::test_perf[llama_v3.2_1b-bench-pytorch-bfloat16-input_output_len:500,2000-reqs:10-con:1-gpus:2]
+  #Mistral-Small-3.1-24B-Instruct-2503
+  - perf/test_perf.py::test_perf[mistral_small_v3.1_24b-bench-pytorch-bfloat16-maxbs:1-input_output_len:1000,2000-reqs:8-con:1-gpus:2]
+  - perf/test_perf.py::test_perf[mistral_small_v3.1_24b-bench-pytorch-bfloat16-input_output_len:1000,2000-reqs:500-con:200-gpus:2]
+  - perf/test_perf.py::test_perf[mistral_small_v3.1_24b-bench-pytorch-bfloat16-maxbs:1-maxnt:20000-input_output_len:20000,2000-reqs:8-con:1-gpus:2] TIMEOUT(120)
+  - perf/test_perf.py::test_perf[mistral_small_v3.1_24b-bench-pytorch-bfloat16-maxbs:4096-maxnt:20000-input_output_len:20000,2000-reqs:500-con:200-gpus:2] TIMEOUT(120)
 
 - condition:
     ranges:
@@ -312,7 +322,6 @@ llm_perf_full:
   - perf/test_perf.py::test_perf[llama_v4_scout_17b_16e_instruct-bench-pytorch-streaming-bfloat16-input_output_len:128,128-ep:8-tp:8-gpus:8]
   - perf/test_perf.py::test_perf[llama_v4_scout_17b_16e_instruct-bench-pytorch-bfloat16-input_output_len:500,2000-ep:8-tp:8-gpus:8]
   - perf/test_perf.py::test_perf[llama_v4_scout_17b_16e_instruct-bench-pytorch-bfloat16-input_output_len:2000,500-ep:8-tp:8-gpus:8]
-  - 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(45)
   #deepseek_r1_fp8
   #pytorch backend
   - perf/test_perf.py::test_perf[deepseek_r1_fp8-bench-pytorch-float8-maxbs:32-input_output_len:128,128-ep:8-tp:8-gpus:8]