 Yanhong Li
|
a911f4dd20
|
[Model] Add support for OLMo Hybrid (#32550)
|
2026-03-05 14:51:06 -05:00 |
|
|
Xin Yang
|
f917020983
|
[Perf] Optimize FusedMoEModularKernel output tensor using torch.empty (#35794)
Signed-off-by: Xin Yang <xyangx@amazon.com>
|
2026-03-05 13:47:53 -05:00 |
|
|
tomeras91
|
86483ca774
|
[Bugfix] Disable FlashInfer TRTLLM BF16 path for non-gated MoE (#36146)
Signed-off-by: Tomer Asida <57313761+tomeras91@users.noreply.github.com>
|
2026-03-05 09:49:05 -08:00 |
|
|
Netanel Haber
|
b93a9e6f6d
|
ParakeetProjection.norm = RMSNorm instead of nn.LayerNorm (#36133)
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
|
2026-03-05 17:29:30 +00:00 |
|
|
Avery Miao
|
e998fa76b9
|
[BUGFIX]Fix Qwen-Omni models audio max_token_per_item estimation error leading to encoder_cache_size is 0 (#35994)
Signed-off-by: Miao, Avery <avery.miao@intel.com>
|
2026-03-05 09:16:29 -08:00 |
|
|
Jiayi YanandHarry Mellor
|
6a895197fa
|
[Bugfix][CI] fix typos (#34934)
Signed-off-by: 1195343015 <1195343015@qq.com>
Signed-off-by: Jiayi Yan <66017932+1195343015@users.noreply.github.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
|
2026-03-05 17:05:46 +00:00 |
|
|
|
3ee68590c7
|
refactor funasr model. (#36108)
Signed-off-by: zixiao <shunli.dsl@alibaba-inc.com>
Co-authored-by: zixiao <shunli.dsl@alibaba-inc.com>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
|
2026-03-05 08:07:37 -08:00 |
|
|
Cyrus Leung
|
7196348157
|
[Bugfix] Fix Qwen-VL tokenizer implementation (#36140)
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
|
2026-03-05 08:07:19 -08:00 |
|
|
Harry Mellor
|
ecde7af9c4
|
Fix import that was moved in Transformers 5.2.0 (#36120)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
|
2026-03-05 13:59:44 +00:00 |
|
|
Kunshang Ji
|
66a2209645
|
[Hardware] Replace torch.cuda.synchronize() api with torch.accelerator.synchronize (#36085)
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
|
2026-03-05 10:36:39 +00:00 |
|
|
Hanjun Choandwang.yuqi
|
f600d5192e
|
[Bugfix] Fix score layer quantization for sequence classification models - Qwen3 (VL) Reranker (#35849)
Signed-off-by: Hanjun Cho <gkswns0531@gmail.com>
Co-authored-by: wang.yuqi <yuqi.wang@daocloud.io>
|
2026-03-04 20:57:20 -08:00 |
|
|
Andrii SkliarandAndrii Skliar
|
0a12cea25f
|
Order config.py in Lexicographical order (#35866)
Signed-off-by: Andrii Skliar <askliar@nvidia.com>
Co-authored-by: Andrii Skliar <askliar@nvidia.com>
|
2026-03-04 20:56:47 -08:00 |
|
|
daje0601andClaude Opus 4.5
|
3b23d57c96
|
[Model] Add LoRA support for Whisper models (#29856)
Signed-off-by: daje0601 <englishmt4118@gmail.com>
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
|
2026-03-05 10:38:25 +08:00 |
|
|
Shanshan Shen
|
562339abc3
|
[Misc] Support OOT linear method registering (#35981)
Signed-off-by: shen-shanshan <467638484@qq.com>
|
2026-03-04 22:25:56 +00:00 |
|
|
amitz-nv
|
d7adcadb9b
|
[Bugfix] Fix passing of activation_type to trtllm fused MoE NVFP4 and FP8 (#36017)
Signed-off-by: amitz-nv <203509407+amitz-nv@users.noreply.github.com>
|
2026-03-04 22:23:51 +00:00 |
|
|
Robert ShawandRobert Shaw
|
b7d59ffce2
|
[UX] Remove NoOpOffloader log (#35678)
Signed-off-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <robshaw@redhat.com>
|
2026-03-04 12:13:40 -08:00 |
|
|
tomeras91
|
7faba503c4
|
[Kernel][Mamba] Optimize Mamba2 SSD prefill Triton kernels (#35397)
Signed-off-by: Tomer Asida <57313761+tomeras91@users.noreply.github.com>
|
2026-03-04 19:47:17 +01:00 |
|
|
tc-mbandhezhihui
|
bfdb512f11
|
fix minicpmo4.5: fix attn_mask in vit attn && fix resampler pos_emb i… (#34127)
Signed-off-by: tc-mb <caitianchi@modelbest.cn>
Co-authored-by: hezhihui <hezhihui@modelbest.cn>
|
2026-03-04 17:46:17 +00:00 |
|
|
Yan Ma
|
58cfe0dc44
|
Fix phi4-mm and remove cuda binding (#35964)
Signed-off-by: Yan Ma <yan.ma@intel.com>
|
2026-03-05 01:08:05 +08:00 |
|
|
Netanel Haber
|
289fc48ab7
|
Use MMEncoderAttention (=use FlashAttention) instead of torch.sdpa in radio.py (#35653)
|
2026-03-04 08:43:13 -08:00 |
|
|
|
6cb901093f
|
[Core] Add All-to-All communication backend for DCP (#34883)
Signed-off-by: Sungsoo Ha <sungsooh@nvidia.com>
Signed-off-by: sungsoo ha <hasungsoo@gmail.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
|
2026-03-04 10:01:57 -05:00 |
|
|
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 |
|
|
Kunshang JiandHarry Mellor
|
16d2ad1d38
|
[Hardware] Replace torch.cuda.empty_cache with torch.accelerator.empty_cache (#30681)
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
Signed-off-by: Kunshang Ji <jikunshang95@gmail.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
|
2026-03-04 09:49:47 +00:00 |
|
|
Chuan (Richard) Li
|
5dc3538736
|
[ROCm][Bugfix] Fall back from CK MXFP4 MoE when GEMM dimensions are unsupported (#35893)
Signed-off-by: Li <chuali@amd.com>
|
2026-03-04 08:30:54 +00:00 |
|
|
Nathan PriceandIsotr0py
|
36bf213181
|
[Bugfix] Add missing dynamic_arg_dims for Qwen3-ASR torch.compile (#35869)
Signed-off-by: Nathan Price <nathan@abridge.com>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
|
2026-03-04 08:29:01 +00:00 |
|
|
+2
|
5d199ac8f2
|
Support Audio Extraction from MP4 Video for Nemotron Nano VL (#35539)
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
Signed-off-by: Andrii Skliar <askliar@nvidia.com>
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Signed-off-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
Signed-off-by: Andrii <askliar@nvidia.com>
Co-authored-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
Co-authored-by: Andrii Skliar <askliar@oci-nrt-cs-001-vscode-01.cm.cluster>
Co-authored-by: Andrii <askliar@nvidia.com>
Co-authored-by: root <root@pool0-03748.cm.cluster>
Co-authored-by: Roger Wang <hey@rogerw.io>
Co-authored-by: root <root@pool0-02416.cm.cluster>
Co-authored-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
Co-authored-by: Matthew Bonanni <mbonanni@redhat.com>
Co-authored-by: Tyler Michael Smith <tyler@neuralmagic.com>
Co-authored-by: wangxiyuan <wangxiyuan1007@gmail.com>
Co-authored-by: root <root@pool0-04880.cm.cluster>
|
2026-03-03 23:20:33 -08:00 |
|
|
Andreas Karatzas
|
edba15045a
|
[Bugfix] Guard mm_token_type_ids kwarg in get_mrope_input_positions (#35711)
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
|
2026-03-04 04:12:51 +00:00 |
|
|
Isotr0py
|
6e9f21e8a2
|
[Chore] Remove debug code in model implementation (#35883)
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
|
2026-03-03 19:50:58 -08:00 |
|
|
|
c1d963403c
|
[model] support FireRedASR2 (#35727)
Signed-off-by: zixiao <shunli.dsl@alibaba-inc.com>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: zixiao <shunli.dsl@alibaba-inc.com>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
|
2026-03-03 19:41:30 -08:00 |
|
|
Shanshan Shen
|
77e6dcbbfa
|
[PluggableLayer][MM] Add PluggableLayer for RelPosAttention (#33753)
Signed-off-by: shen-shanshan <467638484@qq.com>
|
2026-03-03 19:41:27 -08:00 |
|
|
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 |
|
|
xjxandIsotr0py
|
9a9d442464
|
Enable bnb for multiple indices weight (#35838)
Signed-off-by: xjx <493337577@qq.com>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
|
2026-03-04 01:46:47 +00:00 |
|
|
bnellnmandRobert Shaw
|
a9b8b13e5c
|
[Bugfix] Fix misnamed parameter in compressed_tensors_moe.py (#35813)
Signed-off-by: Bill Nell <bnell@redhat.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
|
2026-03-03 16:29:57 -05:00 |
|
|
|
97995f6376
|
[MoE Refactor] Create MK for TRTLLM Kernels (#32564)
Signed-off-by: Robert Shaw <robshaw@redhat.com>
Signed-off-by: Robert Shaw <rshaw@neuralmagic.com>
Signed-off-by: Robert Shaw <robertgshaw2@gmail.com>
Co-authored-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <rshaw@neuralmagic.com>
|
2026-03-03 10:39:50 -08:00 |
|
|
Lucas WilkinsonandMatthew Bonanni
|
28ef9ba399
|
[BugFix] Add support for MTP num_speculative_tokens > 1 with sparse MLA (#34552)
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Co-authored-by: Matthew Bonanni <mbonanni@redhat.com>
|
2026-03-03 07:21:57 -08:00 |
|
|
lin-shh
|
35a6f0bfe2
|
[Misc] Fix typos in comments: explict→explicit, paramaters→parameters (#35648)
|
2026-03-02 21:59:14 -08:00 |
|
|
Isotr0py
|
8ea8ba275e
|
[V0 deprecation] Remove Swin model (#35821)
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
|
2026-03-02 20:03:41 -08:00 |
|
|
Jakub Zakrzewski
|
c8b678e53e
|
[Model] Add support for nvidia/llama-nemotron-rerank-vl-1b-v2 (#35735)
Signed-off-by: Jakub Zakrzewski <jzakrzewski@nvidia.com>
|
2026-03-03 08:32:14 +08:00 |
|
|
Robert ShawandRobert Shaw
|
9319044ee9
|
[MoE][Perf] Wrap DSV3 QKVAProj GEMM in custom op for torch.compile (#35751)
Signed-off-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <robshaw@redhat.com>
|
2026-03-02 23:03:49 +00:00 |
|
|
Ye (Charlotte) Qi
|
fa6a6be519
|
[Bugfix] Fix missing sequence_lengths in qwen3_omni_moe_thinker (#35741)
Signed-off-by: Ye (Charlotte) Qi <yeq@meta.com>
|
2026-03-02 21:11:56 +00:00 |
|
|
Fynn Schmitt-Ulms
|
9433acb8df
|
[Spec Decode] Add hidden states extraction system (#33736)
Signed-off-by: Fynn Schmitt-Ulms <fschmitt@redhat.com>
|
2026-03-02 14:29:09 -05:00 |
|
|
CSWYF3634076
|
2a9e3347e9
|
[BugFix][Model]Fix the garbled code in Ernie4.5-VL caused by fast_moe_cold_start (#35587)
Signed-off-by: wangyafeng <wangyafeng@baidu.com>
|
2026-03-02 18:56:33 +00:00 |
|
|
Turner Jabbour
|
4034c3d32e
|
[Core] Move test utility to test file (#35672)
Signed-off-by: Turner Jabbour <doubleujabbour@gmail.com>
|
2026-03-02 10:56:03 -05:00 |
|
|
|
d9c7730877
|
[Performance] Extract kv update ops from MLA attention backends (#34627)
Signed-off-by: ElizaWszola <ewszola@redhat.com>
Signed-off-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Di Wu <dw2761@nyu.edu>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
|
2026-03-02 10:43:19 -05:00 |
|
|
Kunshang Ji
|
92f5d0f070
|
[XPU] fix mxfp4 activation type (#35691)
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
|
2026-03-02 11:48:39 +08:00 |
|
|
|
8b5014d3dd
|
[Attention] FA4 integration (#32974)
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Signed-off-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
Co-authored-by: Matthew Bonanni <mbonanni@redhat.com>
Co-authored-by: Tyler Michael Smith <tyler@neuralmagic.com>
|
2026-03-01 23:44:57 +00:00 |
|
|
Richard Zou
|
e82fbeec7b
|
[torch.compile] Undo the fast_moe_cold_start hack in torch>=2.11 (#35475)
Signed-off-by: Richard Zou <zou3519@gmail.com>
|
2026-03-01 21:44:22 +00:00 |
|
|
haosdent
|
6290470843
|
[Bugfix] Fix dtype mismatch in RMSNormGated.forward_native() during torch.compile (#35256)
Signed-off-by: haosdent <haosdent@gmail.com>
|
2026-03-01 15:14:46 -05:00 |
|
|
Seungho Yoon
|
5a435507d8
|
fix(mxfp4): return is_monolithic=False when LoRA is enabled for Triton backend (#35382)
Signed-off-by: Seungho Yoon <yoonsnowdev@gmail.com>
|
2026-03-01 09:59:30 -05:00 |
|
|
Asaf Gardin
|
bbf81f9a92
|
[Mamba1] - Kernel Level Chunk Alignment for Prefix Caching (#34798)
Signed-off-by: Josephasafg <ajgard7@gmail.com>
|
2026-03-01 20:40:23 +08:00 |
|