 milesial
|
257015d5e5
|
[MoE] Triton MoE Perf regression - restore low latency path (#39016)
|
2026-04-21 02:37:11 -04:00 |
|
|
Chauncey
|
18563f2072
|
[Misc] Reduce attention logging levels (#40086)
Signed-off-by: chaunceyjiang <chaunceyjiang@gmail.com>
|
2026-04-21 02:09:25 +00:00 |
|
|
bnellnmandRobert Shaw
|
6867bcd076
|
[Bugfix] Replace code that disabled shared expert overlap (#39222)
Signed-off-by: Bill Nell <bnell@redhat.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
|
2026-04-20 19:36:16 -04:00 |
|
|
bai
|
191e3fdaa1
|
Update flashinfer to 0.6.8 (#39959)
Signed-off-by: bai <v@gor.io>
|
2026-04-20 10:37:23 -07:00 |
|
|
bnellnmandRobert Shaw
|
726efe177b
|
[MoE Refactor] Move the shared/fused expert output sum into MoERunnerBase (#35949)
Signed-off-by: Bill Nell <bnell@redhat.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
|
2026-04-20 12:28:46 -04:00 |
|
|
Yan Ma
|
595562651a
|
[XPU] fix MoE triton backend in online fp8 quantization (#40109)
Signed-off-by: Yan Ma <yan.ma@intel.com>
|
2026-04-20 11:31:39 -04:00 |
|
|
Vasiliy Kuznetsov
|
38fa87caca
|
mxfp8 online quant move to new frontend (#40152)
Signed-off-by: Vasiliy Kuznetsov <vasiliy@meta.com>
|
2026-04-20 06:26:12 -07:00 |
|
|
Tao He
|
8936118134
|
[Qwen][Bugfix] Fixes sigmoid activation in torch impl of RMSNormGated. (#40245)
Signed-off-by: Tao He <linzhu.ht@alibaba-inc.com>
|
2026-04-20 04:28:19 +00:00 |
|
|
Shinichi Hemmiandgemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
|
fcb31c1ac3
|
[Bugfix] Properly initialize PerTensorScaleParameter for fused-on-disk checkpoints (#39765)
Signed-off-by: Hemmi Shinichi <shemmi@preferred.jp>
Signed-off-by: Shinichi Hemmi <50256998+Alnusjaponica@users.noreply.github.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
|
2026-04-20 02:53:04 +00:00 |
|
|
danisereb
|
d1135a5087
|
Fix MoE backend selection for LoRA (unquantized MoE) (#40273)
Signed-off-by: Daniel Serebrenik <daserebrenik@nvidia.com>
|
2026-04-19 17:18:40 +00:00 |
|
|
Dan Alistarh
|
ed0622e3a8
|
[Attention] TurboQuant: remove redundant random signs, add prior art attribution (#40194)
Signed-off-by: Dan Alistarh <d.alistarh@gmail.com>
|
2026-04-18 14:31:59 -04:00 |
|
|
Yusuf Mohammad
|
b5f6c5f834
|
Added general ND x ND matmul and unit test for it (#39909)
Signed-off-by: Yusuf <yusufmohammad@live.com>
|
2026-04-18 10:05:21 -04:00 |
|
|
Michael Goin
|
a8bffaa133
|
[Kernel] Add MXFP4 W4A4 CUTLASS MoE kernel for SM100 (#37463)
Signed-off-by: mgoin <mgoin64@gmail.com>
|
2026-04-17 16:42:32 -07:00 |
|
|
sychen52
|
6b2b7bd0eb
|
Add nvfp4 support to reshape_and_cache_flash (#37332)
Signed-off-by: Shiyang Chen <shiychen@nvidia.com>
|
2026-04-17 07:28:00 -07:00 |
|
|
Asaf Gardin
|
ad2b1277f9
|
[Quantization] Consolidate experts_int8 with fp8 online quantization (#38463)
Signed-off-by: Josephasafg <ajgard7@gmail.com>
|
2026-04-16 13:12:20 -07:00 |
|
|
roikoren755
|
b897f00c9c
|
Gate SSU dispatch setup (#40039)
Signed-off-by: Roi Koren <roik@nvidia.com>
|
2026-04-16 13:06:01 -07:00 |
|
|
Fadi Arafehandmergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
445b7093fd
|
[perf][cpu] Accelerate BF16 GELU with LUT impl on Arm CPUs (#37469)
Signed-off-by: Fadi Arafeh <fadi.arafeh@arm.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
2026-04-15 22:26:17 -07:00 |
|
|
vllmellm
|
5f7fab881a
|
[ROCm][FEAT] Integrate aiter gemm w8a8 ptpc (#33773)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
|
2026-04-16 09:55:29 +08:00 |
|
|
zofia
|
fc701c8058
|
[XPU][MXFP4] add mxfp4 quant op for XPU (#39857)
Signed-off-by: Zhu, Zufang <zufang.zhu@intel.com>
|
2026-04-15 12:28:19 +00:00 |
|
|
Zhenzhong Xu
|
60995c05b4
|
[Quantization][Autoround][CPU] Add W4A16 Support (#38192)
Signed-off-by: Zhenzhong1 <zhenzhong.xu@intel.com>
Signed-off-by: Zhenzhong Xu <zhenzhong.xu@intel.com>
|
2026-04-15 18:38:31 +08:00 |
|
|
|
f4b42df048
|
[Attention Backend] TurboQuant: 2-bit KV cache compression with 4x capacity (#38479)
Signed-off-by: vibhavagarwal5 <vibhavagarwal5@gmail.com>
Signed-off-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Xinyu Chen <xinyu1.chen@intel.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
|
2026-04-14 19:57:13 -07:00 |
|
|
Michael GoinandClaude
|
65b9808960
|
[Bugfix] Disable FlashInfer CUTLASS MoE on SM121 (DGX Spark) (#39825)
Signed-off-by: mgoin <mgoin64@gmail.com>
Co-authored-by: Claude <noreply@anthropic.com>
|
2026-04-14 16:03:57 -07:00 |
|
|
bnellnm
|
19ec9a0a62
|
[MoE Refactor] Refactor ZeroExpertFusedMoE into new framework (#35549)
Signed-off-by: Bill Nell <bnell@redhat.com>
|
2026-04-14 16:11:20 -04:00 |
|
|
Jackmin801andRobert Shaw
|
1a9353bb02
|
[MoE] Move GPT OSS Triton kernel experts into fused_moe/experts/ (#39007)
Signed-off-by: Robert Shaw <robertgshaw2@gmail.com>
Signed-off-by: Jackmin801 <ongjackm@gmail.com>
Co-authored-by: Robert Shaw <robertgshaw2@gmail.com>
|
2026-04-14 19:27:39 +00:00 |
|
|
roikoren755
|
ecf5ff7ce3
|
[Mamba] Flashinfer selective_state_update (#36162)
Signed-off-by: Roi Koren <roik@nvidia.com>
|
2026-04-14 15:10:58 -04:00 |
|
|
danielafrimiandroot
|
240f2636ca
|
[Kernel] Support TRTLLM GEN NVFP4 MoE for non-512-aligned hidden dims via weight padding (#39510)
Signed-off-by: root <root@lyris0017.lyris.clusters.nvidia.com>
Signed-off-by: Daniel Afrimi <dafrimi@nvidia.com>
Co-authored-by: root <root@lyris0017.lyris.clusters.nvidia.com>
|
2026-04-14 11:49:56 -07:00 |
|
|
|
dc8df110bc
|
add warning when FP8 KV cache misses prefill query quantization (#39752)
Signed-off-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Albert Cheng (Engrg-Hardware 1) <albecheng@login-lyris02.lyris.clusters.nvidia.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
|
2026-04-14 14:43:05 -04:00 |
|
|
Hexiang Wang
|
f02b3269e7
|
[PluggableLayer][3/N] Apply PluggableLayer to moe-related layers. (#33556)
Signed-off-by: whx-sjtu <2952154980@qq.com>
|
2026-04-14 09:55:00 -04:00 |
|
|
bnellnmandRobert Shaw
|
e1e318af01
|
[MoE Refactor] Remove MoE DP chunking (#39107)
Signed-off-by: Bill Nell <bnell@redhat.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
|
2026-04-14 09:48:05 -04:00 |
|
|
fxmarty-amd
|
4eee77b877
|
[fix][MOE] Fix MOE experts intermediate_size dimension not being narrowed before weight loading (#39688)
Signed-off-by: Felix Marty <Felix.Marty@amd.com>
|
2026-04-14 09:35:28 +00:00 |
|
|
|
8213e8f880
|
Bug/test eagle dp v0 (#38938)
Signed-off-by: Monishver Chandrasekaran <monishverchandrasekaran@gmail.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
Co-authored-by: Nicolò Lucchesi <nlucches@redhat.com>
|
2026-04-13 20:50:08 +00:00 |
|
|
Tyler Michael SmithandClaude Opus 4.6
|
f72b20976c
|
[Bugfix] Reject non-nvfp4 dtypes when using the flashinfer_nvlink_one_sided all2all backend (#39717)
Signed-off-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
|
2026-04-13 19:13:51 +00:00 |
|
|
Yongye Zhu
|
739e5945dc
|
[Quantization] [Refactor] Create special "GptOssMxfp4MoeMethod" (#39604)
Signed-off-by: Yongye Zhu <zyy1102000@gmail.com>
|
2026-04-13 12:53:58 -04:00 |
|
|
Santino Ramos
|
4d042ed85f
|
[Bugfix] Fix tensor shape mismatch in sparse attention with speculative decoding (#39542)
Signed-off-by: Santino Ramos <santinor@inferact.ai>
|
2026-04-13 08:57:38 -07:00 |
|
|
Yi Liu
|
d8ddb31644
|
[Bugfix][CT] Fix KV cache scale handling (#39418)
Signed-off-by: yiliu30 <yi4.liu@intel.com>
|
2026-04-13 10:50:16 -04:00 |
|
|
|
fa6ae31177
|
feat: rename logit_bias/logit_scale to logit_mean/logit_sigma for affine score calibration (#39530)
Signed-off-by: Jesus Federico <jefp@amazon.com>
Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: wang.yuqi <yuqi.wang@daocloud.io>
|
2026-04-13 04:43:44 +00:00 |
|
|
Le Yang
|
21fab0a3db
|
fix(moe): fix RoutedExpertsCapturer assertion failure with DP>1 and MK path (#37879)
|
2026-04-12 10:28:17 -04:00 |
|
|
EdalatiAli
|
df1e30e74b
|
[Quant] add CompressedTensorsW8A8Mxfp8 for linear and MoE layers (#38815)
Signed-off-by: EdalatiAli <aliedalati@cohere.com>
|
2026-04-11 17:21:36 -06:00 |
|
|
|
d4cb783c10
|
[Bugfix] Fix GDN FLA kernel crashes with NULL_BLOCK_ID=0 CUDA graph padding (#39064)
Signed-off-by: Vibhav Agarwal <vibhavagarwal5@gmail.com>
Co-authored-by: vibhav-agarwal <vibhav.agarwal@glance.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
|
2026-04-11 08:35:19 +00:00 |
|
|
Michael Goin
|
11e2375fe2
|
[Refactor] Move MXFP8 GEMM management into MxFp8LinearKernel (#39205)
Signed-off-by: mgoin <mgoin64@gmail.com>
|
2026-04-10 14:02:03 -07:00 |
|
|
Manu
|
51cfc0e76c
|
perf(moe): add tuned fused_moe config for RTX PRO 6000 Blackwell Server Edition (#39183)
Signed-off-by: manu <fortin.emmanuel@gmail.com>
|
2026-04-10 11:32:42 -06:00 |
|
|
Jesus FedericoandClaude Opus 4.6
|
b87575d24b
|
feat: add logit_scale to PoolerConfig for affine score calibration (#39435)
Signed-off-by: Jesus Federico <jefp@amazon.com>
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
|
2026-04-10 17:21:14 +00:00 |
|
|
Jee Jee Li
|
ecd1ea1363
|
[Kernel] Porting the TRTLLM minimax_allreduce_rms kernels (#37045)
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
|
2026-04-11 00:20:20 +08:00 |
|
|
wang.yuqi
|
cb5f7501cb
|
[New Model]: jinaai/jina-reranker-v3 (#38800)
Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
|
2026-04-10 15:20:40 +00:00 |
|
|
Hexiang Wangandmergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
1dfd64c1cc
|
[PluggableLayer][3/N] Apply PluggableLayer to llm_head and vocab embedding layer (#33465)
Signed-off-by: whx-sjtu <2952154980@qq.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
2026-04-10 16:12:59 +08:00 |
|
|
Richard Zou
|
f44afef6d6
|
[compile] Allow strings in custom ops without regressing compilation times (#38123)
Signed-off-by: Richard Zou <zou3519@gmail.com>
|
2026-04-10 07:26:37 +00:00 |
|
|
Kunshang Ji
|
55d037e2e5
|
[CT][FP8][Marlin] refactor CompressedTensorsW8A16Fp8 to use kernel abstraction (#38244)
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
Signed-off-by: Kunshang Ji <jikunshang95@gmail.com>
|
2026-04-10 09:58:35 +08:00 |
|
|
Ibrahim Arshad
|
9853a3c159
|
fix(gdn): Align prefill warmup with real prefill path (#39169)
Signed-off-by: Ibrahim Arshad <38925737+ibrahim1023@users.noreply.github.com>
|
2026-04-10 00:49:50 +00:00 |
|
|
Michael Goin
|
2800706f06
|
[Refactor] Move NVFP4 GEMM management into NvFp4LinearKernel (#39129)
Signed-off-by: mgoin <mgoin64@gmail.com>
|
2026-04-09 15:05:36 -04:00 |
|
|
PikaPikachu
|
827268e98d
|
[Quantization] Support Quark W8A8 INT8 MoE inference (#36320)
Signed-off-by: kangletian <Letian.Kang@amd.com>
|
2026-04-09 17:24:43 +00:00 |
|