Skip to content

Pull requests: flashinfer-ai/flashinfer

New pull request New
94 Open 1,753 Closed
94 Open 1,753 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

Add test case for Qwen3N
#2532 opened Feb 10, 2026 by samuellees Loading…
3 of 5 tasks
4
Add parallel testing to unit test script
#2531 opened Feb 9, 2026 by dierksen Loading…
2 tasks
21
pick fa2 for BatchDecodeWithPagedKVCacheWrapper auto backend
#2530 opened Feb 9, 2026 by saltyminty Loading…
2 tasks
@saltyminty
21
Chore: Cute dsl moe update (TMA.RED implementation)
#2529 opened Feb 9, 2026 by nv-yunzheq Loading…
5 tasks
14
Update Docker CI tags to 20260209-a2d3b39 automated docker
#2528 opened Feb 9, 2026 by flashinfer-bot Loading…
4
feat: BF16 GEMM benchmarking support
#2525 opened Feb 9, 2026 by raayandhar Loading…
5 tasks done
12
Revert "ci: refactor PR tests to hide failed spot jobs from PR status…
#2524 opened Feb 8, 2026 by yongwww Loading…
5 tasks
4
Feat/gdn decode pooled
#2521 opened Feb 8, 2026 by xutizhou Loading…
4 of 10 tasks
8
Support NVFP4 KV cache decode on SM120
#2520 opened Feb 8, 2026 by Tom-Zheng Loading…
5 tasks done
10
perf: avoid redundant shared writes in RMSNorm reductions
#2519 opened Feb 7, 2026 by JackeyLove1 Loading…
5 tasks
4
Improve small size performance in cutedsl fp4
#2517 opened Feb 7, 2026 by vincentzed • Draft
5 tasks
4
benchmarks: Add microbenchmark support for Mamba selective_state_update
#2512 opened Feb 6, 2026 by bkryu Loading…
5 tasks
17
ci: add cleanup step to nightly release self-hosted runner jobs
#2510 opened Feb 6, 2026 by yongwww Loading…
4 of 5 tasks
3
bugfix: fix the enum/int type mismatch mentioned in #2507
#2508 opened Feb 6, 2026 by yzh119 Loading…
4 of 5 tasks
5
mxfp8 trtllm integration
#2505 opened Feb 6, 2026 by IwakuraRein • Draft
5 tasks
17
Ameyn/gdn decode cutedsl kernel
#2498 opened Feb 5, 2026 by ameynaik-hub Loading…
5 tasks done
43
fix: update block size and tile size heuristic for 16+ num of token
#2492 opened Feb 4, 2026 by nv-yunzheq Loading…
5 tasks
6
[WIP][bugfix]Correct chunk_end calculation in multi-CTA collaboration when max_len > length
#2489 opened Feb 4, 2026 by huangzhilin-hzl Loading…
5 tasks
6
feat: Add valid_hidden_size and valid_intermediate_size params t…
#2482 opened Feb 3, 2026 by aleozlx • Draft
3 of 5 tasks
5
feat: Add TRTLLM-Gen Skip-Softmax kernels for prefill and decode run-ci
#2477 opened Feb 3, 2026 by DomBrown Loading…
5 tasks done
27
Proposing a fork owner credential authorization
#2472 opened Feb 3, 2026 by aleozlx • Draft
5 tasks
5
[wip] RoPE kernels: refactor cuda backend to cute dsl
#2470 opened Feb 3, 2026 by kahyunnam • Draft
5 tasks
@kahyunnam
11
ci: Add CUDA 13.1 CI container support
#2465 opened Feb 2, 2026 by bkryu • Draft
5 tasks done
15
feat: Add MXFP8 GEMM mm_mxfp8 (cutlass)
#2464 opened Feb 2, 2026 by danisereb Loading…
5 tasks done
@dhiraj113
59
fix: RMSNorm/FusedRMSNorm + Quant kernels cuda graph fixes
#2459 opened Feb 1, 2026 by BLaZeKiLL Loading…
5 tasks done
9
ProTip! no:milestone will show everything without a milestone.