tzhouamandClaude Fable 5.1 49a839ecf9 [Edge][WP5] Add codec-token streaming mode with on-device decoding ...

* qwen3_tts_talker_only pipeline (single terminal LLM_AR stage,
  final_output_type "latent") + vllm_omni/deploy/qwen3_tts_talker_only.yaml;
  a terminal talker publishes its per-step codes.audio rows to the client
  (omni_client_multimodal_output_keys).
* /v1/audio/speech/stream: session.config.output_mode="codec_tokens" streams
  codec.start (JSON) + binary frames (<IHB seq/n_frames/flags + int16 rows,
  vllm_omni/entrypoints/openai/codec_stream.py) + codec.done; new input.cancel
  message aborts an in-flight generation for both codec and PCM sessions
  (single reader task + session inbox replaces direct socket reads); optional
  session.config.seed; audio.done gains "cancelled" only when it happened.
* serving_speech._generate_codec_chunks aligns the cumulative code tensor to
  output token ids (drops prefill placeholder/EOS rows) and emits deltas.
* TTS adapters: supports_codec_stream / codec_stream_spec (Qwen3-TTS: 16
  codebooks, 2048 ids, 12.5 Hz, 24 kHz, decoder qwen3_tts_tokenizer_v2_12hz).
* examples/online_serving/text_to_speech/qwen3_tts/edge_decoder_client.py
  decodes locally with Qwen3TTSTokenizerV2Decoder (ramp schedule, 72-frame
  context, cancel/reset), codec_parity_check.py and
  benchmarks/tts/codec_stream_e2e.py compare against the two-stage server.
* docs/serving/codec_token_streaming.md; CPU unit tests for the wire format,
  alignment, session config and handler (frames, unsupported adapter, cancel);
  L3 e2e tests/e2e/online_serving/test_qwen3_tts_codec_stream.py.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

2026-09-09 11:10:13 +08:00

..

helpers

[Core] Graduate MiniCPM-o 4.5 and PersonaPlex full-duplex serving out of experimental (#6196)

2026-09-05 08:20:21 +00:00

minimax_h3

[Test] Add MiniMax-H3 DLO DP2 variants and Turbo LoRA L3 test (#6556)

2026-09-01 10:48:52 +00:00

__init__.py

[CI] Qwen3-Omni online test (#257)

2025-12-12 14:25:52 +08:00

nemotron_voicechat_realtime_duplex.py

[Perf][NemotronVoiceChat] Fast defaults for all three deploy profiles: 2.8x offline, realtime full-duplex audio (#6354)

2026-09-03 18:49:19 -07:00

personaplex_realtime_duplex.py

[Core] Graduate MiniCPM-o 4.5 and PersonaPlex full-duplex serving out of experimental (#6196)

2026-09-05 08:20:21 +00:00

run_minicpmo_realtime_duplex_demo_pair.py

[Full Duplex] Feat: Support Full-Duplex realtime runtime & add MiniCPM-o 4.5 demo (#3907)

2026-07-23 16:45:18 +08:00

run_minicpmo_realtime_duplex_multi_session.py

[Core] Graduate MiniCPM-o 4.5 and PersonaPlex full-duplex serving out of experimental (#6196)

2026-09-05 08:20:21 +00:00

run_minicpmo_realtime_duplex_server_vad.py

[Bugfix][MiniCPM-o] Isolate duplex handoff failures and barge-in (#6170)

2026-08-28 15:46:41 +08:00

run_minicpmo_realtime_duplex_soft_interrupt.py

[Bugfix][MiniCPM-o] Anchor duplex soft-interrupt listen sandwich before commit (#6772)

2026-08-30 03:25:43 +00:00

test_audex_tta.py

[New Model] support nemotron Audex (#4976)

2026-07-28 03:58:22 +00:00

test_audex_tts.py

[New Model] support nemotron Audex (#4976)

2026-07-28 03:58:22 +00:00

test_bagel_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_bagel_multi_replicas.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_bagel.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_boogu_image_edit_expansion.py

[Feature][Diffusion]: CFG parallelism support for Boogu-Image (#6786)

2026-09-03 10:37:21 +08:00

test_boogu_image_edit.py

[Diffusion] Add Boogu Edit-Turbo support docs and tests (#6701)

2026-09-01 08:41:23 +08:00

test_boogu_image_expansion.py

[Feature][Diffusion]: CFG parallelism support for Boogu-Image (#6786)

2026-09-03 10:37:21 +08:00

test_boogu_image.py

[CI]demote slow nightly cases and trim Qwen-Image-Edit coverage (#5944)

2026-08-17 11:21:28 +08:00

test_cosmos3.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_cosyvoice3_tts_expansion.py

[Model] Add native full-duplex Nemotron VoiceChat serving (#6089)

2026-08-26 13:19:20 -07:00

test_covo_audio_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_dots_tts_expansion.py

[Feat][dots.tts] Add Online Serving Support (#6235)

2026-09-04 14:44:10 -07:00

test_dreamzero_expansion.py

[Refactor][CI] Reorganize test tree, and move slow e2e to weekly (#4951)

2026-07-29 23:53:03 +08:00

test_duplex_client_live.py

[Core] Graduate MiniCPM-o 4.5 and PersonaPlex full-duplex serving out of experimental (#6196)

2026-09-05 08:20:21 +00:00

test_dynin_omni_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_ernie_image_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_flux2_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_flux_2_dev_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_flux_kontext_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_glm_image_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_glm_tts_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_gr00t_openpi_expansion.py

[Refactor][CI] Reorganize test tree, and move slow e2e to weekly (#4951)

2026-07-29 23:53:03 +08:00

test_hidream_i1_full_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_higgs_audio_v2_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_higgs_audio_v3.py

[BugFix][Higgs-Audio-V3] Fix transcript and concurrent sampling tests (#6422)

2026-08-30 08:52:15 +00:00

test_hunyuan_video_15_expansion.py

Add HYvideo1.5 benchmark (#6349)

2026-08-28 14:31:03 +08:00

test_indextts2_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_krea2_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_lance.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_lingbot_video_expansion.py

[diffusion][feature] Add LingBot-Video T2I and TI2V generation modes (#5311)

2026-08-10 10:02:31 +08:00

test_lingbot_video_moe.py

[CI]demote slow nightly cases and trim Qwen-Image-Edit coverage (#5944)

2026-08-17 11:21:28 +08:00

test_lingbot_video.py

[diffusion][feature] Add LingBot-Video T2I and TI2V generation modes (#5311)

2026-08-10 10:02:31 +08:00

test_lingbot_world_v2_stepwise.py

[Feature][LingBot World]stepwise execution (#6844)

2026-09-07 13:38:23 +08:00

test_longcat_image_edit_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_longcat_image_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_ltx25.py

[Diffusion] Add LTX-2.5 Diffusion VAE decoder support (#6189)

2026-08-27 23:54:40 +08:00

test_ltx.py

[CI]demote slow nightly cases and trim Qwen-Image-Edit coverage (#5944)

2026-08-17 11:21:28 +08:00

test_magi2.py

[Model] Add native MAGI-2 Preview diffusion support (#5918)

2026-09-05 15:32:30 +08:00

test_mimo_audio_expansion.py

[CI]Split GPU jobs by cards_* and reject hand-written SKU marks (#6650)

2026-08-29 18:42:33 +08:00

test_ming_flash_omni_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_ming_tts_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_minicpmo_4_5_duplex_expansion.py

[Bugfix][MiniCPM-o] Isolate duplex handoff failures and barge-in (#6170)

2026-08-28 15:46:41 +08:00

test_minicpmo_4_5_duplex.py

[Core] Graduate MiniCPM-o 4.5 and PersonaPlex full-duplex serving out of experimental (#6196)

2026-09-05 08:20:21 +00:00

test_minicpmo_4_5_expansion.py

[CI] Select mirror_hardwares presets via MIRROR_HW (H100/B200) (#5543)

2026-09-02 15:39:49 +00:00

test_minicpmo_4_5.py

[CI] Select mirror_hardwares presets via MIRROR_HW (H100/B200) (#5543)

2026-09-02 15:39:49 +00:00

test_minicpmo_realtime_duplex_drivers.py

[Core] Graduate MiniCPM-o 4.5 and PersonaPlex full-duplex serving out of experimental (#6196)

2026-09-05 08:20:21 +00:00

test_minimax_h3_dlo_dp2_t2va.py

[diffusion][feature] Add component-selective offload policies (#5929)

2026-09-06 05:40:09 +08:00

test_minimax_music3.py

[Model] Add MiniMax Music 3 text-to-music (#6186)

2026-08-18 04:33:13 +00:00

test_moss_tts_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_moss_tts_nano_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_nemotron_voicechat_duplex.py

[CI] Select mirror_hardwares presets via MIRROR_HW (H100/B200) (#5543)

2026-09-02 15:39:49 +00:00

test_nextstep_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_omnivoice_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_pi0_expansion.py

Add π0 VLA model support (#4222)

2026-08-17 14:29:17 +08:00

test_qwen2_5_omni_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_qwen3_omni_expansion.py

[CI] Select mirror_hardwares presets via MIRROR_HW (H100/B200) (#5543)

2026-09-02 15:39:49 +00:00

test_qwen3_omni_multi_replicas.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_qwen3_omni.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_qwen3_tts_base_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_qwen3_tts_base.py

[CI][Bugfix]Drop dummy-weight Qwen3-TTS Base from Ready CI (#6861)

2026-08-31 12:13:26 +00:00

test_qwen3_tts_codec_stream.py

[Edge][WP5] Add codec-token streaming mode with on-device decoding

2026-09-09 11:10:13 +08:00

test_qwen3_tts_customvoice_expansion.py

[Frontend] Add configurable output sample rate to Speech API (#6553)

2026-09-01 17:25:09 +08:00

test_qwen3_tts_customvoice.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_qwen_image_edit_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_qwen_image_edit.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_qwen_image_expansion.py

[CI] Reduce Qwen-Image Function and share step-execution perf server (#6613)

2026-08-25 22:55:15 +08:00

test_qwen_image_layered_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_qwen_image_layered.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_qwen_image.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_sana_video_expansion.py

[Model] Add native SANA-Video 2B T2V and I2V support (#5508)

2026-08-25 09:46:02 +08:00

test_sana_wm.py

[Model] Add SANA-WM support(Stage-1 only) (#4061)

2026-08-18 23:57:55 -06:00

test_sd3_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_sdxl.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_stable_audio_online_expansion.py

[Misc] move text-to-audio online examples to a unified folder (#4807)

2026-08-25 22:02:26 +08:00

test_step_audio2_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_voxcpm2_tts_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_voxcpm2_tts.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_voxtral_tts_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_wan22_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_wan22_t2v.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_wan_2_1_vace_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_z_image.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00

test_zimage_expansion.py

[Test] Cleanup for helpers (#6523)

2026-08-25 09:10:17 +08:00