Yueqian Lin ede3dddfdf fix(serving): scope no-async-chunk FINAL_ONLY streaming to qwen3_tts ...

async_chunk defaults to False in OmniModelConfig, so coercing every
async_chunk=False deploy to FINAL_ONLY for streaming requests would
silently disable incremental delivery for non-Qwen3 models that stream
DELTA-style today. Gate the coercion on _tts_model_type == "qwen3_tts"
(the model this PR validated) and pin the non-qwen3 behavior with a test.

Signed-off-by: Yueqian Lin <linyueqian@outlook.com>

2026-07-03 22:57:27 -07:00

..

__init__.py

[Feature] Support standard OpenAI API sampling parameters for thinker stage. (max_tokens...) (#450)

2025-12-24 20:39:01 +08:00

conftest_video.py

[Feature] Streaming video input with EVS frame filtering (RFC #2201 Phase 2-4) (#2342)

2026-04-24 22:54:35 +08:00

test_decode_audio_url.py

Revert "[Bugfix] Use MediaConnector for image/video URL fetching to prevent SSRF" (#4751)

2026-06-27 01:18:48 +08:00

test_image_server.py

[Bugfix] Qwen-Image-Layered: allow layers=2 to match upstream demo (#4777)

2026-07-04 03:06:24 +00:00

test_omni_sleep_wakeup.py

[Bugfix] fix when diffusion model not set sleeping_stages (#3023)

2026-05-25 23:13:38 +08:00

test_openai_audio_list_endpoints.py

[CI] Reducing mithril-h100-pool in ready&merge CI, optimize ready CI duration (#4354)

2026-06-29 05:17:57 +00:00

test_openpi_connection.py

[Perf][DreamZero]dreamzero supports CUDA graph, torch.compile, and DiT Caching. (#4213)

2026-06-19 14:24:37 +08:00

test_openpi_serving.py

Support DROID policy server for Cosmos3 OpenPI (#4282)

2026-06-15 21:22:59 +08:00

test_qwen3_omni_realtime_websocket.py

[CI] Reducing mithril-h100-pool in ready&merge CI, optimize ready CI duration (#4354)

2026-06-29 05:17:57 +00:00

test_qwen3_tts_speaker_embedding.py

Make speech streaming default to SSE (#4679)

2026-06-30 02:17:24 +00:00

test_qwen3_tts_websocket.py

[Bugfix] Stream Qwen3-TTS WebSocket input as one request (#4731)

2026-06-26 07:48:30 +00:00

test_serving_audio_generate.py

[Frontend]Handle audio generate engine errors consistently (#3316)

2026-05-17 14:10:26 +00:00

test_serving_chat_metrics.py

[BugFix] Surface diffusion metrics in chat completions; sanitize Bagel img2img mm kwargs (#2932)

2026-04-22 16:57:01 +00:00

test_serving_chat_multistage_generation.py

[Model] Aura support - Non async chunk path (#4257)

2026-06-21 22:08:26 +08:00

test_serving_chat_sampling_params.py

[Model] Aura support - Non async chunk path (#4257)

2026-06-21 22:08:26 +08:00

test_serving_chat_speaker.py

[Rebase] Rebase to vllm v0.24.0 (#4709)

2026-06-30 19:41:06 +08:00

test_serving_speech_batch.py

[CI] Reducing mithril-h100-pool in ready&merge CI, optimize ready CI duration (#4354)

2026-06-29 05:17:57 +00:00

test_serving_speech_stream.py

[Bugfix] Stream Qwen3-TTS WebSocket input as one request (#4731)

2026-06-26 07:48:30 +00:00

test_serving_speech.py

fix(serving): scope no-async-chunk FINAL_ONLY streaming to qwen3_tts

2026-07-03 22:57:27 -07:00

test_serving_video_output_stream.py

[feat] Streaming diffusion video generation output (#3737)

2026-06-23 09:57:53 +00:00

test_serving_video_stream.py

Refactor: extract OmniStreamingVideoHandler base and QwenOmniStreamingVideoHandler (#4424)

2026-06-17 15:07:45 +00:00

test_speech_usage.py

[Bugfix] Fix /v1/audio/speech usage token accounting for Qwen3-TTS (#4646) (#4673)

2026-06-27 06:13:31 +08:00

test_stage_params.py

[Refactor] Centralize stage sampling params resolution (#3153)

2026-04-26 23:30:11 +08:00

test_storage_manager.py

[Frontend] Configurable video storage backends with TTL (#2531)

2026-06-19 17:41:22 -06:00

test_tts_adapter.py

[Refactor] TTS serving adapter framework + migrate TTS models (excl. ming_flash_omni) (#4330)

2026-06-13 04:22:54 +00:00

test_video_api_utils.py

[feat] Streaming diffusion video generation output (#3737)

2026-06-23 09:57:53 +00:00

test_video_frame_filter.py

[Feature] Streaming video input with EVS frame filtering (RFC #2201 Phase 2-4) (#2342)

2026-04-24 22:54:35 +08:00

test_video_server.py

Revert "[Bugfix] Use MediaConnector for image/video URL fetching to prevent SSRF" (#4751)

2026-06-27 01:18:48 +08:00

test_video_stream_handler.py

[Rebase] Rebase to vllm 0.20.0 (#3232)

2026-04-30 09:56:59 +12:00

test_video_stream_session.py

[Feature] Streaming video input with EVS frame filtering (RFC #2201 Phase 2-4) (#2342)

2026-04-24 22:54:35 +08:00