anodyine/vllm-omni
1
0 Fork 0
Code Issues 4 Pull Requests 2 Actions Packages Projects Releases Wiki Activity
957 Commits 235 Branches 31 Tags
034d579d770bb39fa2fb8d2bd1caca27f1affae3
Commit Graph
609 Commits
Author SHA1 Message Date
tzhouam 034d579d77 Update vLLM precompiled wheel version and refactor various components for improved functionality and error handling. Changes include modifying the get_name method in DiffusionQuantizationConfig to use class-level attributes, enhancing metadata handling in StageEngineCoreClient, and adjusting the AsyncOmni class constructor for better parameter management. Additionally, improved error handling in omni_snapshot_download and optimized initial chunk size computation in talker2code2wav_async_chunk for dynamic load adaptation. 
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
 2026-03-19 05:14:34 +00:00
tzhouam 4f75cccacc Merge remote-tracking branch 'origin/main' into dev/vllm-align 2026-03-19 03:11:40 +00:00
Joshna-Medisettyandgcanlin ca6c7ad2be Fix OmniGen2 transformer config loading for HF models (#1934) 
Signed-off-by: Joshna Medisetty <joshna.medisetty@intel.com>
Signed-off-by: gcanlin <canlinguosdu@gmail.com>
Co-authored-by: gcanlin <canlinguosdu@gmail.com>
 2026-03-19 10:21:54 +08:00
tzhouam 342a209924 Merge remote-tracking branch 'origin/main' into dev/vllm-align 
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
 2026-03-19 02:06:26 +00:00
TJian c85acb1773 [CI] [ROCm] Bugfix device environment issue (#1984) 
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
 2026-03-19 05:11:28 +08:00
Juan Pablo ZuluagapabloHongsheng Liu
 		fc5e8f825e [Optim][Qwen3TTS] big boost model throughput+latency high concurrency (#1852) 
Signed-off-by: pablo <pablo@agigo.ai>
Signed-off-by: JuanPZuluaga <juanz9312@gmail.com>
Co-authored-by: pablo <pablo@agigo.ai>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
 2026-03-18 12:22:05 -04:00
Taichang Zhou dc6f2fb81f Enhance cu_seqlens computation in Qwen3Omni_VisionTransformer by switching to tensor operations for improved performance and compatibility. Added error handling to fallback on a vectorized implementation if necessary. This update aligns with upstream changes and optimizes the handling of sequence lengths during attention execution. 2026-03-18 23:27:55 +08:00
Taichang Zhou 7a68bd37b7 Refactor cu_seqlens handling in Qwen3Omni_VisionTransformer to align with upstream vLLM changes. The sequence length helper now retains cu_seqlens in numpy format and converts it for attention execution, improving compatibility and performance. 2026-03-18 23:23:41 +08:00
SYLAR 53ebdd0b1d [Bug]: fix CUDA OOM during diffusion post-processing (#1670) 
Signed-off-by: lishunyang <lishunyang12@163.com>
 2026-03-18 14:59:58 +00:00
汪志鹏 61e170c69a [Feature]: Remove some useless hf_overrides in yaml (#1898) 
Signed-off-by: princepride <wangzhipeng628@gmail.com>
 2026-03-18 19:46:44 +08:00
Y. FisherandCanlin Guo f52b51533a [Feature] Add cache-dit support for HunyuanImage3 (#1848) 
Signed-off-by: KexiongYu <yukexiong1@huawei.com>
Co-authored-by: Canlin Guo <canlinguosdu@gmail.com>
 2026-03-18 10:18:27 +00:00
Canlin Guo d636c32ff9 [Bugfix] Set PREEMPTED status when moving requests from running to waiting queue (#1893) 
Signed-off-by: gcanlin <canlinguosdu@gmail.com>
 2026-03-18 17:56:13 +08:00
Chenguang ZhengyinpePeiqi YinwuhangPeiqi Yinlinyueqianprincepridelishunyang
 		3158e9d28f [Entrypoint][Refactor] vLLM-Omni Entrypoint Refactoring (#1908) 
Signed-off-by: yinpe <11810305@mail.sustech.edu.cn>
Signed-off-by: Peiqi Yin <yinpeiqi809@gmail.com>
Signed-off-by: yinpeiqi <yinpeiqi809@gmail.com>
Signed-off-by: wuhang <whlbx@hotmail.com>
Signed-off-by: linyueqian <linyueqian@outlook.com>
Signed-off-by: princepride <wangzhipeng628@gmail.com>
Signed-off-by: lishunyang <lishunyang12@163.com>
Signed-off-by: Chenguang ZHENG <645327136@qq.com>
Signed-off-by: Peiqi Yin <60515999+yinpeiqi@users.noreply.github.com>
Co-authored-by: yinpe <11810305@mail.sustech.edu.cn>
Co-authored-by: Peiqi Yin <yinpeiqi809@gmail.com>
Co-authored-by: wuhang <whlbx@hotmail.com>
Co-authored-by: Peiqi Yin <60515999+yinpeiqi@users.noreply.github.com>
Co-authored-by: linyueqian <linyueqian@outlook.com>
Co-authored-by: princepride <wangzhipeng628@gmail.com>
Co-authored-by: lishunyang <lishunyang12@163.com>
 2026-03-18 17:33:53 +08:00
Yueqian Lin fe786f1a19 Fix Base voice clone streaming quality and stop-token crash (#1945) 
Signed-off-by: linyueqian <linyueqian@outlook.com>
 2026-03-18 06:54:12 +08:00
Taichang Zhou 8a31e7865d Increase timeout for Diffusion Model Test and Diffusion Images API LoRA E2E in Buildkite configurations to 30 minutes. Also, introduce a new timeout constant for diffusion initialization in relevant test files to enhance stability during testing. Update OmniServer initialization to utilize the new timeout settings. 2026-03-17 23:43:46 +08:00
tzhouam 004c60a031 Refactor shutdown and cleanup logic in OmniRunner and OmniStage to improve process termination and resource management. The changes ensure that stage workers are stopped before closing queues, preventing potential orphaned subprocesses. Additionally, enhanced error handling during process termination provides clearer logging for failed shutdown attempts. This update aims to increase the reliability and stability of the shutdown process across the application. 
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
 2026-03-17 09:33:28 +00:00
tzhouam cc4ade4ba7 Refactor shutdown logic in OmniLLM and stage worker to ensure proper engine closure. The shutdown method now checks for the presence of engine_core before invoking shutdown, enhancing resource management. Additionally, the stage worker now explicitly calls stage_engine.close() upon receiving a shutdown signal. This improves the overall stability and reliability of the shutdown process. 
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
 2026-03-17 08:19:56 +00:00
tzhouam 722ad6bbae rebase: align vllm-omni with vLLM 0a0a1a198be8 
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
 2026-03-17 03:26:42 +00:00
tzhouam 0eccb77c4f Merge remote-tracking branch 'origin/main' into dev/vllm-align 2026-03-17 02:48:50 +00:00
Juan Pablo Zuluaga cf328eb8e8 [Bug][Qwen3TTS][Streaming] remove dynamic initial chunk and only compute on initial request (#1930) 
Signed-off-by: JuanPZuluaga <juanz9312@gmail.com>
 2026-03-16 15:23:09 -04:00
rongfu.leng 4036cd547c [Misc] removed qwen3_tts.py as it is out-dated (#1926) 
Signed-off-by: rongfu.leng <lenronfu@gmail.com>
 2026-03-16 13:09:04 -04:00
tzhouam de6bcc734c Update Dockerfile.ci to use a new vllm wheel URL and refactor MooncakeConnector import logic in monkey_patch.py for improved compatibility. Adjust OpenAIServingChat to call the harmony request method from the correct context. 
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
 2026-03-16 08:33:28 +00:00
tzhouam a26e568dfc Merge remote-tracking branch 'origin/main' into dev/vllm-align 2026-03-16 07:30:43 +00:00
Canlin Guo 9718a9e369 [NPU] Upgrade to v0.17.0 (#1890) 
Signed-off-by: gcanlin <canlinguosdu@gmail.com>
 2026-03-16 14:42:59 +08:00
tzhouam 9c22e3afb0 Refactor import statement for OpenAIServingEmbedding in api_server.py to use aliasing for clarity 
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
 2026-03-16 06:13:07 +00:00
Jinheng 88caaf182b [Feature] Split #1303 Part 1: PD disaggregation scaffolding (#1863) 
Signed-off-by: Jinheng Li <ahengljh@gmail.com>
 2026-03-16 14:05:19 +08:00
tzhouam aae101f495 Refactor import statement for OpenAIServingEmbedding in api_server.py 
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
 2026-03-16 03:39:51 +00:00
tzhouam 3a508d7b7a rebase: align vllm-omni with vLLM e9163b536e72 
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
 2026-03-16 02:32:39 +00:00
tzhouam f137a1e3ad Merge remote-tracking branch 'origin/main' into dev/vllm-align 2026-03-16 01:49:46 +00:00
Lancer 9aae75c3ee [Feat] support HSDP for Flux family (#1900) 
Signed-off-by: Lancer <maruixiang6688@gmail.com>
 2026-03-15 22:48:12 +08:00
Lancer 2472c360aa [Feat] support SP for FLUX.2-klein (#1250) 
Signed-off-by: Lancer <maruixiang6688@gmail.com>
 2026-03-15 12:18:12 +08:00
Lancer af012babf8 [Bugfix] Fix cpu offload and quantization compatibility (#1473) 
Signed-off-by: Lancer <maruixiang6688@gmail.com>
 2026-03-14 18:55:30 +00:00
SYLARGao HanHongsheng Liu
 		5377619c0b fix: propagate parallel_config through create_default_diffusion (#1878) 
Signed-off-by: lishunyang <lishunyang12@163.com>
Signed-off-by: Hongsheng Liu <liuhongsheng4@huawei.com>
Co-authored-by: Gao Han <hgaoaf@connect.ust.hk>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
 2026-03-14 17:07:13 +08:00
Alex Brooks e17130090c [BugFix] Fix Max Rank Handling in LoRA (#1397) 
Signed-off-by: Alex Brooks <albrooks@redhat.com>
 2026-03-14 02:42:54 +00:00
tzhouam 97899c6c98 Merge remote-tracking branch 'origin/main' into dev/vllm-align 2026-03-13 19:03:57 +00:00
Yueqian Lin cebf78ba3e [Bugfix] Restore voice upload API and profiler endpoints reverted by #1719 (#1879) 
Signed-off-by: linyueqian <linyueqian@outlook.com>
 2026-03-14 04:37:08 +13:00
dengyunyang 3e8cb6c631 [Diffusion] [Model] Dreamid omni (#1855) 
Signed-off-by: dengyunyang <584797741@qq.com>
 2026-03-13 21:53:06 +08:00
Ian EavesSamitGao Han
 		92f771d455 [Frontend] Rewrite video API for async job lifecycle (#1665) 
Signed-off-by: Ian Eaves <ian.k.eaves@gmail.com>
Signed-off-by: Samit <285365963@qq.com>
Co-authored-by: Samit <285365963@qq.com>
Co-authored-by: Gao Han <gaohan19@huawei.com>
 2026-03-13 17:11:06 +08:00
汪志鹏 e0b71056ba [BugFix]: Fix bagel online inference bug (#1804) 
Signed-off-by: princepride <wangzhipeng628@gmail.com>
 2026-03-13 08:58:44 +00:00
Lancer 53a3df7309 [Bugfix] fix layer-wise offload incompatible with cache-dit (#1786) 
Signed-off-by: Lancer <maruixiang6688@gmail.com>
 2026-03-13 15:28:13 +08:00
SYLAR 75056b2db2 [Bugfix] Fix SD3.5-medium attn2 uninitialized weights (#1659) 
Signed-off-by: lishunyang <lishunyang12@163.com>
 2026-03-13 14:58:17 +08:00
tzhouam d3db476f2d Merge branch 'main' into dev/vllm-align 2026-03-13 06:51:22 +00:00
tzhouam 62c8695dfd Update GPUGenerationModelRunner to clarify KV connector metadata handling during draft model runs 
- Changed variable name to better reflect its purpose in managing KV connector metadata.
- Updated comments for improved clarity regarding the handling of speculative configurations.

Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
 2026-03-13 03:16:33 +00:00
Alex BrooksandDidan Deng c8cf8a799d [Refactor] Use SP Plan for LongCat Sequence Parallelism (#1772) 
Signed-off-by: Alex Brooks <albrooks@redhat.com>
Co-authored-by: Didan Deng <33117903+wtomin@users.noreply.github.com>
 2026-03-13 11:13:34 +08:00
tzhouam d911475b22 Fix output queue handling in OmniStage to return None if queue is closed during shutdown 
Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
 2026-03-13 01:59:04 +00:00
tzhouam e17fb0e97e Refactor OpenAI serving components and enhance multimodal handling 
- Updated import for OpenAIServingEmbedding in api_server.py.
- Enhanced omni_init_app_state to initialize renderer for engine_client.
- Added handle_oov_mm_token parameter to multiple model classes for better multimodal token handling.
- Improved comments for clarity in various model files.
- Adjusted GPUARModelRunner to ensure proper handling of late interaction runner attributes.

Signed-off-by: [Your Name] <[Your Email]>

Signed-off-by: tzhouam <tzhouam@connect.ust.hk>
 2026-03-13 01:55:17 +00:00
Yueqian Lin 366b3368ea Add Fish Speech S2 Pro support with online serving and voice cloning (#1798) 
Signed-off-by: linyueqian <linyueqian@outlook.com>
 2026-03-12 16:20:52 -04:00
Yueqian Lin 9c6fbd76be Enable async_scheduling by default for Qwen3-TTS (#1853) 
Signed-off-by: linyueqian <linyueqian@outlook.com>
 2026-03-12 12:30:20 -04:00
Du Bin 8ad8988e2b [Bugfix] Fix Helios text_encoder embed_tokens all-zeros due to untied weights (#1728) 
Signed-off-by: OSS Scout <scout@oss-scout.local>
 2026-03-12 15:28:21 +00:00
usberkeleyprinceprideHongsheng Liu
 		b3cced310a [bugfix] Add Interleaved 2D Rotary Embedding for HunyuanImage3 (#1784) 
Signed-off-by: Bradley <bradley.b.pitt@gmail.com>
Signed-off-by: princepride <wangzhipeng628@gmail.com>
Signed-off-by: 汪志鹏 <wangzhipeng628@gmail.com>
Co-authored-by: princepride <wangzhipeng628@gmail.com>
Co-authored-by: Hongsheng Liu <liuhongsheng4@huawei.com>
 2026-03-12 21:33:03 +08:00