44 Commits
Author SHA1 Message Date
SeongJun Leeandlesj0610 a98247ab4d [Model] Drop the 4x M-RoPE cache headroom for Qwen3-VL, Qwen3.5 and Qwen3.8 (#59853)
Signed-off-by: lesj0610 <lesj0610@users.noreply.github.com>
Signed-off-by: lesj0610 <lesj0610@gmail.com>
Co-authored-by: lesj0610 <lesj0610@users.noreply.github.com>
2026-10-10 06:51:52 +00:00
cjackalandmergify[bot] <37929162+mergify[bot]@users.noreply.github.com> 491f44adfa [Model] Extend device-side mm normalization to Qwen3VL/Qwen3.5/Qwen4Next (#51289)
Signed-off-by: cjackal <44624812+cjackal@users.noreply.github.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2026-09-29 11:44:54 +08:00
Ashraf Bhuiyan 5113cf9d16 [Mypy] Fix mypy typing for Qwen and Qianfan models (#58046)
Signed-off-by: Ashraf Bhuiyan <mbhuiyan@redhat.com>
2026-09-26 14:31:51 +00:00
Alexander CrettenandandClaude 7d8d71e989 [Bugfix] Qwen3-VL(-MoE): pass architectures to with_hf_config for pipeline parallelism (#43272)
Signed-off-by: Xonder <xonder@gmail.com>
Co-authored-by: Claude <noreply@anthropic.com>
2026-09-10 14:49:35 +01:00
Harry MellorandClaude Opus 5 80d6d557f3 Standardise weight tying on ParallelLMHead.tie_weights (#52147)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-08-13 15:10:09 +00:00
Andreas Karatzasandmergify[bot] <37929162+mergify[bot]@users.noreply.github.com> 36f4630d89 [Bugfix][MoE] Fix fused block-scale orientation (#50727)
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2026-08-11 16:05:38 -04:00
Wei-Cheng (Wayne) ChiuandOpenAI Codex 7794b1e08b [Bugfix] Skip Qwen3 deepstack buffers without vision (#49397)
Signed-off-by: WEI CHENG CHIU <waynehacking8@gmail.com>
Co-authored-by: OpenAI Codex <noreply@openai.com>
2026-08-05 12:46:23 +08:00
nvbfalk 60915c972c [Feature] Add VidCom2 video token pruning (#47750)
Signed-off-by: Benedikt Falk <bfalk@nvidia.com>
2026-07-28 03:20:17 +00:00
wenpengw-nvandRoger Wang 9a21f0d1a3 [BugFix] Initialize model_config for Qwen3-VL MoE (#44863)
Signed-off-by: wenpengw-nv <wenpengw@nvidia.com>
Co-authored-by: Roger Wang <hey@rogerw.io>
2026-07-13 13:43:53 -07:00
Harry Mellor 1ab9522935 Remove more unnecessary load_weights methods (#47058)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
2026-06-30 15:22:16 +01:00
Harry Mellor d8eb734d94 Fix Transformers backend FP8 MoE and remove some boilerplate (#46820)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
2026-06-27 00:16:05 +01:00
bnellnm f4966f8b3d [Bugfix] Fix weight loading issues caused by #41184 (#45054)
Signed-off-by: Bill Nell <bnell@redhat.com>
2026-06-10 01:20:13 -04:00
weizhoublue a94189295b Fix Weight loading for Qwen3.5-MTP and Qwen3-VL using runai_streamer (#42716)
Signed-off-by: weizhoublue <weizhou.lan@daocloud.io>
2026-05-16 17:54:27 -07:00
SILONG ZENG cee6751e54 [Bugfix][Qwen3-VL] Fix pipeline-parallel deepstack initialization (#42394)
Signed-off-by: MrZ20 <2609716663@qq.com>
2026-05-13 10:58:42 +00:00
ℍ𝕠𝕝𝕝𝕠𝕨 𝕄𝕒𝕟 908a713488 [Bugfix] LoRA: extend expert base_layer loading to Qwen3.5 and Step3.x (#37114)
Signed-off-by: Hollow Man <hollowman@opensuse.org>
2026-04-21 14:17:03 +00:00
Benjamin Chislett 8b346309a5 [Refactor] Consolidate SupportsEagle (#36063)
Signed-off-by: Benjamin Chislett <bchislett@nvidia.com>
2026-03-13 23:22:40 +00:00
Raghavan c8c3935b70 [Bugfix][Model] Fix FP8 k_scale/v_scale not loaded for Qwen3-MoE (#35656)
Signed-off-by: raghavan <oneraghavan@gmail.com>
2026-03-04 13:15:38 +00:00
William Zhang 70c73df69e [Bugfix] Fix EVS implementation for Qwen3 VL (#33607)
Signed-off-by: 2ez4bz <133824995+2ez4bz@users.noreply.github.com>
2026-03-04 02:18:11 +00:00
Harry MellorandRoger Wang 1e9204bff3 Make Qwen3VL compatible with Transformers v5 (#34262)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Signed-off-by: Roger Wang <hey@rogerw.io>
Co-authored-by: Roger Wang <hey@rogerw.io>
2026-02-11 04:13:23 -08:00
Isotr0py 192ad4648b [Bugfix] Fix interns1-pro initialization and PP (#33793)
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
2026-02-04 17:54:45 +00:00
zxyandIsotr0py a3acfa1071 [Models] Intern-S1-Pro (#33636)
Signed-off-by: zxy <zhou0493@e.ntu.edu.sg>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
2026-02-03 05:49:45 -08:00
Cyrus Leung dcd80206b7 [Chore] Update type annotation of input_ids in model forward (#33063)
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
2026-01-26 06:02:10 -08:00
Isotr0py 9ad7f89f55 [Models]: Make Multimodal config implicit in ViT implementation (#31972)
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
2026-01-24 20:34:26 +08:00
Cyrus Leung 2b8a38b6d6 [Model] Extend collect_children and no_init_weights contexts (#32757)
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
2026-01-22 08:20:27 +00:00
shanjiaz 7ab80a8e37 Added qwen3 vision language moe support for speculative decoding (#32048)
Signed-off-by: shanjiaz <zsjwpianpian@gmail.com>
Signed-off-by: shanjiaz <43143795+shanjiaz@users.noreply.github.com>
2026-01-21 03:24:05 +00:00
Cyrus Leung fda3f03eb2 [4/N] Initialize MM components in context managers (M-P) (#32663)
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
2026-01-20 14:06:32 +00:00
Cyrus Leung 4753f3bf69 [Model] Use context managers for encoder- and LM-only mode (#32605)
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
2026-01-20 11:43:38 +08:00
Max Hu 3f175f18a2 [Bugfix] Fix multimodal configuration for Qwen3VL MOE model (#30670)
Signed-off-by: Max Hu <hyoung2991@gmail.com>
2025-12-15 14:06:01 +00:00
87b4d1557d [CustomOp][MM] Extract MMEncoderAttention as CustomOp and replace the backend of QwenVisionAttention with it. (#30125)
Signed-off-by: shen-shanshan <467638484@qq.com>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: tjtanaa <tunjian.tan@embeddedllm.com>
2025-12-15 11:13:32 +08:00
Jee Jee Li 2f5f9acd55 [LoRA] Continue optimizing MoE LoRA weight loading (#29322)
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
2025-11-27 05:56:28 -08:00
JartX 8e38e99829 [Feature] EPLB on Qwen3VLMoe and CompressedTensorsWNA16MoEMethod (#28849) 2025-11-19 18:30:08 -05:00
Harry Mellor 97d1c99302 Rename clashing method names for vLLM model protocol (#27583)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
2025-11-12 19:14:33 -08:00
Jee Jee Li daec4d2624 [Model]Improve Qwen3VLMoeForConditionalGeneration packed_modules_mapping (#27096)
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
2025-10-17 04:47:00 -07:00
Harry Mellor 8fcaaf6a16 Update Optional[x] -> x | None and Union[x, y] to x | y (#26633)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
2025-10-12 09:51:31 -07:00
Lukas Geiger 93f2c0aa08 [Models] Improve iteration over layers (#26425)
Signed-off-by: Lukas Geiger <lukas.geiger94@gmail.com>
2025-10-08 20:48:33 +00:00
Harry Mellor d6953beb91 Convert formatting to use ruff instead of yapf + isort (#26247)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
2025-10-05 07:06:22 -07:00
whx cbf9221992 [Model] Supplement to PR 24862: Pass param prefix to LLMHead (#25805)
Signed-off-by: whx-sjtu <2952154980@qq.com>
2025-10-03 21:34:53 +08:00
Roger Wang 66bca9b8bd [MM] Add text-only mode for Qwen3-VL (#26000) 2025-09-30 21:13:42 -07:00
JJJYmmmandliuye.hj 471997adf6 [Bugfix] fix Qwen3VLMoe load when pp > 1 (#25838)
Signed-off-by: liuye.hj <liuye.hj@alibaba-inc.com>
Co-authored-by: liuye.hj <liuye.hj@alibaba-inc.com>
2025-09-28 17:56:12 +00:00
Isotr0py d4d9899860 [Quantization] Add field to skip unquantized modules for GPTQ config (#25455)
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
2025-09-26 15:47:41 +00:00
Roger Wang be874c0201 [Bugfix] Fix Qwen3-VL-MoE weight loading for EP (#25300)
Signed-off-by: Roger Wang <hey@rogerw.io>
2025-09-20 00:04:05 -07:00
Roger Wang 1dfea5f4a9 [Bugfix][Perf] Misc fixes for Qwen3 VL (#25238)
Signed-off-by: Roger Wang <hey@rogerw.io>
2025-09-19 10:46:16 +00:00
Roger WangHuang Jie松灵gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>Isotr0py
3127274d02 [MM Encoder] Apply DP ViT for Qwen3-VL model series (#24955)
Signed-off-by: Roger Wang <hey@rogerw.io>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Huang Jie <92386084+JJJYmmm@users.noreply.github.com>
Co-authored-by: 松灵 <26085463+wulipc@users.noreply.github.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
2025-09-17 21:04:21 -07:00
0f7acdd73c [Model] Support Qwen3-VL Model Series (#24727)
Signed-off-by: Roger Wang <hey@rogerw.io>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Huang Jie <92386084+JJJYmmm@users.noreply.github.com>
Co-authored-by: 松灵 <26085463+wulipc@users.noreply.github.com>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
2025-09-17 05:01:04 +00:00