Commits · 54254f7a6155002b1a493c3cefda9752bc9ce92f · OpenDAS / vllm_cscc

21 Feb, 2026 3 commits
- [ROCm][CI] Fix spec decode logprobs flakiness and parametrize tree attention backends (#34599) · 54254f7a
  Andreas Karatzas authored Feb 20, 2026
```
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
```
  54254f7a
- [feat] Add per-block extra_keys to KV events (#33304) · a0fe7ea2
  zhongdaor-nv authored Feb 20, 2026
```
Signed-off-by: zhongdaor-nv <zhongdaor@nvidia.com>
Co-authored-by: Nicolò Lucchesi <nlucches@redhat.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
```
  a0fe7ea2
- [Kernel] Optimize sample_recovered_tokens_kernel (#34974) · 7a5adad4
  Xin Yang authored Feb 20, 2026
```
Signed-off-by: Xin Yang <xyangx@amazon.com>
```
  7a5adad4
20 Feb, 2026 3 commits
- [CI] Revert PRs 34818 and 33600 (#34979) · aaefc58e
  Lucas Wilkinson authored Feb 20, 2026
  
  aaefc58e
- [Test] Add FP8 KV Cache Testing for MLA Backends (#34473) · f24b2de3
  Wei Zhao authored Feb 20, 2026
```
Signed-off-by: wzhao18 <wzhao18.sz@gmail.com>
```
  f24b2de3
- [CI][AMD][BugFix][P/D] Add default_vllm_config to test_moriio_connector.py so tests pass (#33739) · 0c1dc427
  rasmith authored Feb 19, 2026
```
Signed-off-by: Randall Smith <Randall.Smith@amd.com>
```
  0c1dc427
19 Feb, 2026 1 commit

[Bugfix] Fix Basic Models Test (#34818) · 662205d3

Matthew Bonanni authored Feb 19, 2026


Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>

662205d3

18 Feb, 2026 1 commit
- [Docs] Clean up speculators docs (#34065) · 64ac1395
  Kyle Sayers authored Feb 18, 2026
```
Signed-off-by: Kyle Sayers <kylesayrs@gmail.com>
```
  64ac1395
17 Feb, 2026 3 commits

[Kernel] Triton-based Top-k and Top-p sampler kernels (#33538) · c656ba3b

Jongseok Park authored Feb 17, 2026


Signed-off-by: js_park <cakeng@naver.com>
Signed-off-by: Jongseok Park <37990712+cakeng@users.noreply.github.com>
Signed-off-by: Sunga Kim <sunga.kim@berkeley.edu>
Signed-off-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: Sunga Kim <sunga.kim@berkeley.edu>
Co-authored-by: Nick Hill <nickhill123@gmail.com>

c656ba3b

[Renderer] Move InputPreprocessor into Renderer (2/2) (#34560) · 574fe752
Cyrus Leung authored Feb 17, 2026
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
574fe752
[CI][BugFix] ShellCheck cleanup to remove baseline and preserve runtime behavior (#34514) · c61a98f5
junuxyz authored Feb 17, 2026
```
Signed-off-by: junuxyz <216036880+junuxyz@users.noreply.github.com>
```
c61a98f5

16 Feb, 2026 1 commit
- [Scheduler][ASR] Fix CrossAttn blocks per-request for Variable length encoder inputs (#31058) · cd81cdb3
  Ekagra Ranjan authored Feb 16, 2026
```
Signed-off-by: Ekagra Ranjan <3116519+ekagra-ranjan@users.noreply.github.com>
Co-authored-by: Nicolò Lucchesi <nlucches@redhat.com>
```
  cd81cdb3
14 Feb, 2026 3 commits
- [Hybrid] Enable mamba prefix cache "align" mode with async scheduling (#33997) · d5fe3f70
  Thomas Parnell authored Feb 14, 2026
```
Signed-off-by: Thomas Parnell <tpa@zurich.ibm.com>
```
  d5fe3f70
- [Renderer] Move InputPreprocessor into Renderer (1/2) (#34510) · 73391a1b
  Cyrus Leung authored Feb 15, 2026
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Signed-off-by: Cyrus Leung <cyrus.tl.leung@gmail.com>
```
  73391a1b
- [ROCm][CI] Guard sparse MLA backend imports for ROCm compatibility in tests (#34538) · b3c14229
  Andreas Karatzas authored Feb 14, 2026
```
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
```
  b3c14229
13 Feb, 2026 5 commits

[Hybrid] Enable spec decoding in mamba cache align mode (#33705) · c027541e
Harry Huang authored Feb 14, 2026
```
Signed-off-by: huanghaoyan.hhy <huanghaoyan.hhy@alibaba-inc.com>
```
c027541e

[Bugfix]: Fix structured output in multi-turn gpt-oss (#34454) · fd267bc7

Ben Browning authored Feb 13, 2026


Signed-off-by: Ben Browning <bbrownin@redhat.com>
Co-authored-by: Cyrus Leung <cyrus.tl.leung@gmail.com>
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>

fd267bc7

[Core] Move pause and resume functions into engine (#34125) · dddbff46

Aaron Hao authored Feb 13, 2026


Signed-off-by: ahao-anyscale <ahao@anyscale.com>
Signed-off-by: Aaron Hao <ahao@anyscale.com>
Signed-off-by: hao-aaron <ahao@anyscale.com>
Signed-off-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: Nick Hill <nickhill123@gmail.com>

dddbff46

[Bug Fix] Fix MambaManager.cache_blocks() crash on null blocks in align mode (#34418) · 4137c5df
haosdent authored Feb 13, 2026
```
Signed-off-by: haosdent <haosdent@gmail.com>
```
4137c5df
[Refactor] Simplify BOS/EOS token handling (#34435) · ea5ff3a1
Cyrus Leung authored Feb 13, 2026
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
ea5ff3a1

12 Feb, 2026 3 commits

[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

[V0 Deprecation] Remove code related to per-request logits processors (#34400) · fb455ed5
Cyrus Leung authored Feb 12, 2026
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
fb455ed5
[Refactor] Pass Renderer to Input Processor (#34329) · b96f7314
Cyrus Leung authored Feb 12, 2026
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
b96f7314

11 Feb, 2026 2 commits
- 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
- [Misc] Clean up validation logic in input processor (#34144) · b5dcb372
  Cyrus Leung authored Feb 11, 2026
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  b5dcb372
10 Feb, 2026 5 commits
- [SM100] Resubmit FMHA FP8 prefill for MLA (#31195) · 578977bb
  Pavani Majety authored Feb 10, 2026
```
Signed-off-by: Pavani Majety <pmajety@nvidia.com>
```
  578977bb
- [Core][BugFix] Fix PP KV cache sharding memory validation (#33698) · c5a66d16
  junuxyz authored Feb 11, 2026
```
Signed-off-by: junuxyz <216036880+junuxyz@users.noreply.github.com>
```
  c5a66d16
- [V1][BugFix] Fix EAGLE3 encoder cache miss with disable_chunked_mm_input (#34220) · 748625cd
  Krish Gupta authored Feb 10, 2026
```
Signed-off-by: KrxGu <krishom70@gmail.com>
```
  748625cd
- [BugFix] Avoid prefix cache hit in the same schedule step for mamba layers (#29387) · 97fa8f65
  Chen Zhang authored Feb 09, 2026
```
Signed-off-by: Chen Zhang <zhangch99@outlook.com>
```
  97fa8f65
- [Bugfix][Core] Fix CPU memory leak from Request reference cycle in prefix caching (#34183) · 8a5e0e2b
  Roger Wang authored Feb 09, 2026
```
Signed-off-by: Roger Wang <hey@rogerw.io>
```
  8a5e0e2b
07 Feb, 2026 2 commits

[Misc] Make `PlaceholderRange.get_num_embeds` a method (#34035) · 48312e57
Cyrus Leung authored Feb 07, 2026
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
48312e57

[Feat][RL] Pause and Resume with keep requests for single engine (#32351) · 89a385d7

Aaron Hao authored Feb 06, 2026


Signed-off-by: ahao-anyscale <ahao@anyscale.com>
Signed-off-by: Aaron Hao <ahao@anyscale.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>

89a385d7

06 Feb, 2026 2 commits
- [KV Connector] Add missing method overrides to MultiConnector (#33292) · aca59674
  Seiji Eicher authored Feb 06, 2026
```
Signed-off-by: Seiji Eicher <seiji@anyscale.com>
```
  aca59674
- [Feature] OTEL tracing during loading (#31162) · 325ab6b0
  emricksini-h authored Feb 06, 2026
  
  325ab6b0
05 Feb, 2026 4 commits
- [Spec Decode] Unified Parallel Drafting (#32887) · af3162d3
  Benjamin Chislett authored Feb 05, 2026
```
Signed-off-by: Benjamin Chislett <bchislett@nvidia.com>
```
  af3162d3
- Enable Cross layers KV cache layout at NIXL Connector V2 (#33339) · 8322d4e4
  liranschour authored Feb 05, 2026
```
Signed-off-by: Liran Schour <lirans@il.ibm.com>
Signed-off-by: liranschour <liranschour@users.noreply.github.com>
Co-authored-by: Or Ozeri <or@ozery.com>
Co-authored-by: Nicolò Lucchesi <nicolo.lucchesi@gmail.com>
Co-authored-by: Nicolò Lucchesi <nlucches@redhat.com>
```
  8322d4e4
- [KV Connector][Metrics] Do not count local prefix cache hits in connector queries (#30522) · 2abd9759
  Mark McLoughlin authored Feb 05, 2026
```
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
```
  2abd9759
- [Minor] Include `StreamingInput` in inputs package (#33856) · add9f1fb
  Nick Hill authored Feb 04, 2026
```
Signed-off-by: Nick Hill <nickhill123@gmail.com>
```
  add9f1fb
04 Feb, 2026 2 commits

[Core] Don't schedule spec tokens with prefill chunks (#33652) · fa4e0fb0
Nick Hill authored Feb 04, 2026
```
Signed-off-by: Nick Hill <nickhill123@gmail.com>
```
fa4e0fb0

[KV Connector][BugFix] scheduler: Delay freeing blocks of aborted async loads (#32255) · 8e326908

Or Ozeri authored Feb 04, 2026



Fixes a not-yet-reported case where it was possible for blocks to be
freed by an abort before an async transfer completed, resulting
in corrupted KV data.
Signed-off-by: Or Ozeri <oro@il.ibm.com>

8e326908