46 lines
1.5 KiB
JSON
46 lines
1.5 KiB
JSON
[
|
|
{
|
|
"test_name": "test_qwen3_omni_stability_async_chunk",
|
|
"server_params": {
|
|
"model": "Qwen/Qwen3-Omni-30B-A3B-Instruct",
|
|
"extra_cli_args": ["--async-chunk"]
|
|
},
|
|
"benchmark_params": [
|
|
{
|
|
"dataset_name": "random-mm",
|
|
"backend": "openai-chat-omni",
|
|
"endpoint": "/v1/chat/completions",
|
|
"duration_sec": 86400,
|
|
"max_concurrency": 2,
|
|
"num_prompts_per_batch": 50,
|
|
"random_input_len": {
|
|
"min": 0,
|
|
"max": 8000
|
|
},
|
|
"random_output_len": {
|
|
"min": 0,
|
|
"max": 1000
|
|
},
|
|
"random_range_ratio": 0.0,
|
|
"random_mm_base_items_per_request": {
|
|
"min": 0,
|
|
"max": 6
|
|
},
|
|
"random_mm_num_mm_items_range_ratio": 0.0,
|
|
"random_mm_limit_mm_per_prompt": {
|
|
"image": 2,
|
|
"video": 2,
|
|
"audio": 2
|
|
},
|
|
"random_mm_bucket_config": {
|
|
"(128-1024, 128-1024, 1)": 0.34,
|
|
"(256-1080, 256-1920, 2-16)": 0.33,
|
|
"(0, 1-60, 1-3)": 0.33
|
|
},
|
|
"ignore_eos": true,
|
|
"percentile-metrics": "ttft,tpot,itl,e2el,audio_rtf,audio_ttfp,audio_duration"
|
|
}
|
|
]
|
|
}
|
|
]
|