Commits · 116f4be405ff8aaadbc885d8d527d1694c7fcf0a · OpenDAS / vllm_cscc

01 Apr, 2026 1 commit
- [1/N][Cleanup] Standardize on use of `is_quantized_kv_cache` (#38659) · 116f4be4
  Matthew Bonanni authored Apr 01, 2026
```
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
```
  116f4be4
30 Mar, 2026 1 commit

[Bugfix] Use null block (0) for padded block table entries (#35431) · bcc6f674

SandishKumarHN authored Mar 30, 2026


Signed-off-by: SandishKumarHN <sandish@fb.com>
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Co-authored-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
Co-authored-by: Matthew Bonanni <mbonanni@redhat.com>

bcc6f674

26 Mar, 2026 2 commits
- [Bugfix] Restore CUDA graph persistent buffers for FP8 FlashMLA decode (#35175) · 0aac2048
  haosdent authored Mar 27, 2026
```
Signed-off-by: haosdent <haosdent@gmail.com>
Co-authored-by: Matthew Bonanni <mbonanni@redhat.com>
```
  0aac2048
- [Revert] Remove DeepGEMM availability check in DeepseekV32IndexerMetadataBuilder (#38076) · 87f05d68
  Chauncey authored Mar 26, 2026
```
Signed-off-by: chaunceyjiang <chaunceyjiang@gmail.com>
```
  87f05d68
25 Mar, 2026 2 commits

[ROCm] Utilize persistent MLA kernel from AITER (#36574) · 978fc18b
Sathish Sanjeevi authored Mar 25, 2026
```
Signed-off-by: Sathish Sanjeevi <sathish.krishnan.p.s@gmail.com>
```
978fc18b

[Revert] Remove CUDA torch fallbacks for... · 09c3dc91

Chauncey authored Mar 25, 2026


[Revert] Remove CUDA torch fallbacks for fp8_mqa_logits/fp8_paged_mqa_logits_torch function (#37968)
Signed-off-by: chaunceyjiang <chaunceyjiang@gmail.com>

09c3dc91

23 Mar, 2026 1 commit

[Bugfix] Register VLLM_BATCH_INVARIANT in envs.py to fix spurious unknown env var warning (#35007) · dc6908ac

Ranran authored Mar 23, 2026


Signed-off-by: Ranran <1012869439@qq.com>
Signed-off-by: Ranran <hzz5361@psu.edu>
Signed-off-by: ran <hzz5361@psu.edu>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>

dc6908ac

20 Mar, 2026 1 commit
- [BugFix] Allow qk_nope_head_dim=192 in FlashInfer MLA backend checks (#37475) · e5ed6c6c
  Kaihang Jiang authored Mar 20, 2026
```
Signed-off-by: Kaihang Jiang <kaihangj@nvidia.com>
```
  e5ed6c6c
19 Mar, 2026 1 commit

[Bugfix] Zero-init MLA attention output buffers to prevent NaN from CUDA graph padding (#37442) · ef2c4f77

Elvir Crnčević authored Mar 19, 2026


Signed-off-by: Elvir Crncevic <elvircrn@gmail.com>
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Co-authored-by: Matthew Bonanni <mbonanni@redhat.com>

ef2c4f77

18 Mar, 2026 1 commit

[Bugfix] Fix KV scales inconsistency in fp8 MLA & FlashInfer kv_cache_dtype... · 577df69b

Andy Lo authored Mar 18, 2026


[Bugfix] Fix KV scales inconsistency in fp8 MLA & FlashInfer kv_cache_dtype "auto" leading to gibberish (#37054)
Signed-off-by: Andy Lo <andy@mistral.ai>

577df69b

16 Mar, 2026 3 commits
- [Misc] Add `float16` to `CacheDType` (#37199) · 93f3c8e5
  Matthew Bonanni authored Mar 16, 2026
```
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
```
  93f3c8e5
- [MTP][Sparse MLA] Take advantage of native MTP support in indexer when possible (#36982) · c88ea833
  Matthew Bonanni authored Mar 16, 2026
```
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
```
  c88ea833
- [Bugfix] Disable cross-layer KV cache for MLA attention backends (#37090) · ca1954d5
  haosdent authored Mar 17, 2026
```
Signed-off-by: haosdent <haosdent@gmail.com>
Co-authored-by: Or Ozeri <oro@il.ibm.com>
```
  ca1954d5
13 Mar, 2026 1 commit

[Feat][Bugfix] Enable additional dimension for Flashinfer MLA and fix routing dtype (#36931) · 367cf5cd

Dimitrios Bariamis authored Mar 14, 2026


Signed-off-by: Dimitrios Bariamis <12195802+dbari@users.noreply.github.com>
Co-authored-by: Dimitrios Bariamis <12195802+dbari@users.noreply.github.com>

367cf5cd

12 Mar, 2026 1 commit
- [Kernel] Add FP8 KV cache support to Triton MLA decode attention (#34597) · a1257fd1
  grimulkan authored Mar 12, 2026
```
Signed-off-by: grimulkan <grimulkan@gmail.com>
```
  a1257fd1
11 Mar, 2026 3 commits
- Add XPU MLA Sparse backend for DeepSeek v3.2 (#33230) · e584dce5
  Wuxun Zhang authored Mar 11, 2026
```
Signed-off-by: Zhang, Wuxun <wuxun.zhang@intel.com>
```
  e584dce5
- AITER MLA backend: Avoid CPU sync in _build_decode (#35765) · eac2dc2b
  pschlan-amd authored Mar 11, 2026
```
Signed-off-by: Patrick Schlangen <pschlan@amd.com>
```
  eac2dc2b
- [DSV3.2][MTP] Optimize Indexer MTP handling (#36723) · 9040cd40
  Benjamin Chislett authored Mar 11, 2026
```
Signed-off-by: Benjamin Chislett <bchislett@nvidia.com>
```
  9040cd40
10 Mar, 2026 2 commits
- [ROCm][Perf] Enable `sparse_mla`'s cudagraph on ROCm platform (#35719) · 82f3f30e
  Pleaplusone authored Mar 11, 2026
```
Signed-off-by: ganyi <ygan@amd.com>
```
  82f3f30e
- [Bugfix][Sparse MLA] report indexer CG support properly (#36519) · 9095cbbf
  Matthew Bonanni authored Mar 10, 2026
```
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
```
  9095cbbf
09 Mar, 2026 3 commits

[ROCm][CI] Fix ROCm attention backend validation for head sizes, block sizes,... · c174d54f

Andreas Karatzas authored Mar 09, 2026


[ROCm][CI] Fix ROCm attention backend validation for head sizes, block sizes, and compute capability checks (#36292)
Signed-off-by: Andreas Karatzas <akaratza@amd.com>

c174d54f

[Attention][Perf][Kernel] Replace torch.cat with vectorized CUDA kernel MLA... · 580864d8

Roberto L. Castro authored Mar 09, 2026


[Attention][Perf][Kernel] Replace torch.cat with vectorized CUDA kernel MLA query concat - DeepSeek-V3.2 (#34917)
Signed-off-by: LopezCastroRoberto <rocastro@redhat.com>
Signed-off-by: Roberto L. Castro <38211239+LopezCastroRoberto@users.noreply.github.com>

580864d8

Reapply [Attention] Refactor `check_and_update_config` (#35122) · 77a73458
Matthew Bonanni authored Mar 09, 2026
```
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
```
77a73458

07 Mar, 2026 1 commit
- [Perf] Support FP8 KV cache for Flashinfer MLA Sparse (#35891) · 379689d5
  Wei Zhao authored Mar 07, 2026
  
  379689d5
06 Mar, 2026 2 commits
- [ROCm] Support MLA with nhead<16 and FP8 KV cache for TP=8 (Kimi K2.5/Linear) (#35850) · c188749b
  Chuan (Richard) Li authored Mar 06, 2026
```
Signed-off-by: Li <chuali@amd.com>
```
  c188749b
- Reenable features for ROCm attention backends (#36185) · c5362c73
  Rohan Potdar authored Mar 05, 2026
```
Signed-off-by: Rohan138 <rohanpotdar138@gmail.com>
```
  c5362c73
05 Mar, 2026 2 commits

[Bugfix][CI] fix typos (#34934) · 6a895197

Jiayi Yan authored Mar 06, 2026

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>

6a895197

[ROCm] Refactor ROCm attention backend selection logic (#35246) · 8c760b6a
Sage Moore authored Mar 05, 2026
```
Signed-off-by: Sage Moore <sage@neuralmagic.com>
```
8c760b6a

03 Mar, 2026 1 commit

[BugFix] Add support for MTP num_speculative_tokens > 1 with sparse MLA (#34552) · 28ef9ba3

Lucas Wilkinson authored Mar 03, 2026


Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Co-authored-by: Matthew Bonanni <mbonanni@redhat.com>

28ef9ba3

28 Feb, 2026 1 commit
- [Feat] Add CUDA torch fallbacks for fp8_mqa_logits/fp8_paged_mqa_logits_torch function (#35271) · 7e08c22b
  Chauncey authored Feb 28, 2026
```
Signed-off-by: chaunceyjiang <chaunceyjiang@gmail.com>
```
  7e08c22b
25 Feb, 2026 1 commit
- [Platform] Add current_platform.num_compute_units interface (#35042) · 8ad54a99
  Kunshang Ji authored Feb 25, 2026
```
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
Signed-off-by: Kunshang Ji <jikunshang95@gmail.com>
```
  8ad54a99
24 Feb, 2026 1 commit
- [Bug][DSV3.2] Always prepare metadata for DeepGEMM Sparse Attention (#35075) · 9fa5b25a
  Benjamin Chislett authored Feb 24, 2026
```
Signed-off-by: Benjamin Chislett <bchislett@nvidia.com>
```
  9fa5b25a
17 Feb, 2026 1 commit
- [Bugfix][MTP][Sparse MLA] Allow sparse MLA with MTP to run with FULL cudagraphs (#34457) · dc5fa77a
  Matthew Bonanni authored Feb 17, 2026
```
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>
```
  dc5fa77a
12 Feb, 2026 1 commit

[Attention] Add FlashInfer Sparse MLA backend (#33451) · f2c47886

Matthew Bonanni authored Feb 12, 2026


Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>

f2c47886

11 Feb, 2026 1 commit
- Reapply [Attention][FA3] Update FA3 to include new swizzle optimization (#34043) · c7914d30
  Lucas Wilkinson authored Feb 11, 2026
```
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
```
  c7914d30
10 Feb, 2026 1 commit

[Perf][Kernel] Add faster topKperRow decode kernel for DeepSeek-V3.2 sparse attention (#33680) · afdce12c

Roberto L. Castro authored Feb 10, 2026

Signed-off-by: LopezCastroRoberto <rocastro@redhat.com>
Signed-off-by: Roberto L. Castro <38211239+LopezCastroRoberto@users.noreply.github.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>

afdce12c

05 Feb, 2026 1 commit
- Revert "[Attention][FA3] Update FA3 to include new swizzle optimization" (#33841) · e3bf79ff
  Luka Govedič authored Feb 04, 2026
  
  e3bf79ff
03 Feb, 2026 1 commit
- [Attention][FA3] Update FA3 to include new swizzle optimization (#23465) · 2267cb1c
  Lucas Wilkinson authored Feb 03, 2026
```
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
```
  2267cb1c
31 Jan, 2026 2 commits
- [Attention] Clarify comment explaining attn_logits +1 dimension (#33427) · 5b55c0be
  Francesco Fusco authored Jan 31, 2026
```
Signed-off-by: Francesco Fusco <ffu@zurich.ibm.com>
```
  5b55c0be
- [Attention] Move MLA `forward` from backend to layer (#33284) · aaa901ad
  Matthew Bonanni authored Jan 30, 2026
```
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
```
  aaa901ad