anodyine/vllm-omni
1
0 Fork 0
Code Issues 4 Pull Requests 2 Actions Packages Projects Releases Wiki Activity

[CI] Harden Qwen3-TTS perf nightly: enable Base voice_clone, add c=64/128, 2-GPU split (#3491)

Browse Source 
Signed-off-by: Yueqian Lin <linyueqian@outlook.com>
This commit is contained in:
Yueqian Lin authored and GitHub committed 2026-05-14 14:26:59 +08:00
1 parent 0d9d57acd9
commit c4f859bf56
6 files changed
+51 -24
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

tests/dfx/perf/tests/test_tts.json
+26 -19
View File
@@ -2,16 +2,19 @@
{
"test_name": "test_qwen3_tts_base",
"server_params": {
"model": "Qwen/Qwen3-TTS-12Hz-1.7B-Base"
"model": "Qwen/Qwen3-TTS-12Hz-1.7B-Base",
"stage_overrides": {
"1": {"devices": "1"}
}
},
"benchmark_params": [
{
"task": "voice_clone",
"eval_phase": "latency",
"enabled": false,
"dataset_name": "seed-tts",
"backend": "openai-audio-speech",
"endpoint": "/v1/audio/speech",
"dataset_path": "linyueqian/seed-tts-eval-subset",
"num_prompts": [20],
"max_concurrency": [1],
"seed_tts_locale": "en",
@@ -24,18 +27,18 @@
{
"task": "voice_clone",
"eval_phase": "throughput",
"enabled": false,
"dataset_name": "seed-tts",
"backend": "openai-audio-speech",
"endpoint": "/v1/audio/speech",
"num_prompts": [80],
"max_concurrency": [8],
"dataset_path": "linyueqian/seed-tts-eval-subset",
"num_prompts": [80, 128, 256],
"max_concurrency": [8, 64, 128],
"seed_tts_locale": "en",
"percentile-metrics": "ttft,e2el,audio_rtf,audio_ttfp,audio_duration",
"baseline": {
"median_audio_ttfp_ms": [3500],
"median_audio_rtf": [0.75],
"audio_throughput": [10.0]
"median_audio_ttfp_ms": [3500, 7000, 12000],
"median_audio_rtf": [0.75, 1.5, 2.5],
"audio_throughput": [10.0, 40.0, 50.0]
}
},
{
@@ -45,6 +48,7 @@
"dataset_name": "seed-tts",
"backend": "openai-audio-speech",
"endpoint": "/v1/audio/speech",
"dataset_path": "linyueqian/seed-tts-eval-subset",
"num_prompts": [200],
"max_concurrency": [4],
"seed_tts_locale": "en",
@@ -59,7 +63,10 @@
{
"test_name": "test_qwen3_tts_customvoice",
"server_params": {
"model": "Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice"
"model": "Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice",
"stage_overrides": {
"1": {"devices": "1"}
}
},
"benchmark_params": [
{
@@ -86,15 +93,15 @@
"backend": "openai-audio-speech",
"endpoint": "/v1/audio/speech",
"dataset_path": "benchmarks/build_dataset/seed_tts_smoke",
"num_prompts": [80],
"max_concurrency": [8],
"num_prompts": [80, 128, 256],
"max_concurrency": [8, 64, 128],
"seed_tts_locale": "en",
"extra_body": {"voice": "Vivian", "language": "English", "task_type": "CustomVoice"},
"percentile-metrics": "ttft,e2el,audio_rtf,audio_ttfp,audio_duration",
"baseline": {
"median_audio_ttfp_ms": [1500],
"median_audio_rtf": [0.30],
"audio_throughput": [30.0]
"median_audio_ttfp_ms": [1500, 3000, 5500],
"median_audio_rtf": [0.30, 0.60, 1.0],
"audio_throughput": [30.0, 100.0, 120.0]
}
},
{
@@ -157,15 +164,15 @@
"backend": "openai-audio-speech",
"endpoint": "/v1/audio/speech",
"dataset_path": "benchmarks/build_dataset/seed_tts_design",
"num_prompts": [80],
"max_concurrency": [8],
"num_prompts": [80, 128, 256],
"max_concurrency": [8, 64, 128],
"seed_tts_locale": "en",
"extra_body": {"task_type": "VoiceDesign", "language": "English"},
"percentile-metrics": "ttft,e2el,audio_rtf,audio_ttfp,audio_duration",
"baseline": {
"median_audio_ttfp_ms": [1500],
"median_audio_rtf": [0.35],
"audio_throughput": [25.0]
"median_audio_ttfp_ms": [1500, 3500, 6000],
"median_audio_rtf": [0.35, 0.70, 1.2],
"audio_throughput": [25.0, 80.0, 100.0]
}
},
{