Skip to content

Pull requests: flashinfer-ai/flashinfer

New pull request New
63 Open 1,521 Closed
63 Open 1,521 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

[wip] feat: support variable sequence length in decode kernel of trtllm-gen attention
#2125 opened Nov 20, 2025 by yaoyaoding • Draft
5 tasks
1
[DO NOT MERGE] v0.5.3 release
#2121 opened Nov 20, 2025 by yzh119 Loading…
5 tasks
12
perf: using multi-cta optimization for top-k/top-p
#2119 opened Nov 20, 2025 by yzh119 Loading…
4 of 5 tasks
9
Refactor trtllm_mnnvl_allreduce
#2118 opened Nov 20, 2025 by timlee0212 Loading…
5 tasks done
32
update autotuner input tensor random range
#2116 opened Nov 19, 2025 by jiahanc Loading…
3 of 5 tasks
21
refactor: update fa3 codebase and fix hopper unittest [part 1]
#2111 opened Nov 19, 2025 by yzh119 Loading…
4 of 5 tasks
1
9
feat: support more head dim in RoPE kernel
#2109 opened Nov 19, 2025 by raayandhar Loading…
5 tasks done
14
[wip] feat: Enable API Logging for Better Debugging POC
#2108 opened Nov 18, 2025 by bkryu Loading…
5 tasks
@bkryu
15
enable xqa speculative decoding
#2105 opened Nov 18, 2025 by qsang-nv Loading…
5 tasks
19
Port TRT-LLM communication kernels to flashinfer
#2102 opened Nov 18, 2025 by djns99 • Draft
5 tasks
1
make DeepGEMM swapAB available for linear gemm SM90
#2101 opened Nov 17, 2025 by xuanzic • Draft
5 tasks
1
refactor: pass hopper deepgemm include directory through python
#2090 opened Nov 14, 2025 by yzh119 Loading…
4 of 5 tasks
21
[To merge AFTER flashinfer-ci changes updated] Reduce test time by moving compilation off-line
#2089 opened Nov 14, 2025 by kahyunnam Loading…
5 tasks done
@kahyunnam
15
feat: add sink to flashinfer decode
#2087 opened Nov 13, 2025 by djmmoss Loading…
14
feat: BF16 GEMM using CUTLASS backend for SM100
#2070 opened Nov 10, 2025 by raayandhar Loading…
5 tasks done
43
Add custom communicator for trtllm_mnnvl_ar
#2056 opened Nov 6, 2025 by wenscarl Loading…
5 tasks
7
Rebase FP8 SM100 Cutlass FMHA Attention to main (original PR#1238)
#2047 opened Nov 5, 2025 by pavanimajety • Draft
5 tasks
2
Added an initial implementation of Q and KV Cache in fp8 and to use t…
#2035 opened Nov 4, 2025 by Anerudhan • Draft
3 tasks done
1
Refactor flashinfer/__init__.py so that applications could selectively pack submodules without modifying __init__.py
#2027 opened Nov 3, 2025 by bangshengtang Loading…
5 tasks done
9
Blockwise GEMM with all reduce overlapping
#2007 opened Oct 30, 2025 by Amir-19 • Draft
5 tasks
1
refactor: backend_requirement + supported_compute_capability decorator for gemm
#2000 opened Oct 29, 2025 by jimmyzho Loading…
5 tasks
23
feat: Add backend='auto' to mm_fp4 and enable autotune for backend='cudnn'
#1979 opened Oct 25, 2025 by bkryu Loading…
5 tasks done
@bkryu
134
chore: agentic workflow for automatic version bump
#1947 opened Oct 19, 2025 by yzh119 Loading…
5 tasks
35
add blockwise gemm cute dsl
#1922 opened Oct 13, 2025 by Amir-19 Loading…
5 tasks
33
Sampling non contiguous
#1916 opened Oct 12, 2025 by zcin Loading…
5 tasks done
1
8
ProTip! Find all pull requests that aren't related to any open issues with -linked:issue.