Skip to content

Pull requests: flashinfer-ai/flashinfer

New pull request New
41 Open 929 Closed
41 Open 929 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

Allow cudnn prefill kernels to be called natively
#1317 opened Jul 24, 2025 by Anerudhan • Draft
5 tasks done
14
minor: add trtllm_gen_mla benchmark
#1316 opened Jul 24, 2025 by yyihuang Loading…
5 tasks done
5
[WIP] Add benchmark scripts to different moe gemms
#1315 opened Jul 24, 2025 by fzyzcjy • Draft
5 tasks
8
test qkvo quantization not equal to 1.
#1314 opened Jul 24, 2025 by weireweire Loading…
5 tasks
3
Wrap cudnn backend to unified interface
#1312 opened Jul 23, 2025 by cyx-6 Loading…
5 tasks
5
Support loading autotuned results from json for cutlass fp4 moe backends
#1310 opened Jul 23, 2025 by kaixih Loading…
6
Refactor Fused Moe Module
#1309 opened Jul 23, 2025 by wenscarl Loading…
5 tasks
5
Api regression test for trtllmgen fp8 moe
#1308 opened Jul 23, 2025 by aleozlx Loading…
5 tasks done
6
fix: a workaround to make fp8 kv-cache work for prefill
#1304 opened Jul 22, 2025 by chenyang78 Loading…
2 tasks
11
3rparty: upgrade cutlass dependency to v4.1.0
#1299 opened Jul 22, 2025 by yzh119 Loading…
5 tasks
2
Add weight layout
#1297 opened Jul 21, 2025 by aleozlx Loading…
5 tasks done
7
add mm_fp4 use cutlass backend for large bs
#1296 opened Jul 21, 2025 by ttyio Loading…
5 tasks done
11
[fix] fix integer overflow in FA2 customized_mask & add buffer overflow warning.
#1290 opened Jul 19, 2025 by happierpig Loading…
5 tasks done
10
Add native cudnn_decode for improved cudnn decode performance
#1283 opened Jul 18, 2025 by Anerudhan Loading…
5 tasks done
12
ci: add github actions to upload sdist to pypi
#1270 opened Jul 16, 2025 by yzh119 Loading…
5 tasks
1
Bug fix: fix duplicate launch in POD
#1267 opened Jul 16, 2025 by Edenzzzz Loading…
5 tasks
2
use NVSHMEM4Py instead of custom bindings for NVSHMEM MNNVL Allreduce
#1263 opened Jul 15, 2025 by Amir-19 Loading…
5 tasks
12
feat(aot): add nvshmem module for aot compilation
#1261 opened Jul 15, 2025 by EmilienM Loading…
3 of 5 tasks
1
13
refactor: separate SM100 and legacy TRT-LLM comm modules
#1259 opened Jul 15, 2025 by EmilienM Loading…
3 of 5 tasks
1
14
feat: update cubins for dynamic scale factor interface and one-copy kvcache on trtllm-gen mla
#1248 opened Jul 14, 2025 by yyihuang • Draft
5 tasks done
4
bugfix: fix fp32 acc threshold for qk using math::inf according to dtype by AIDC-AI
#1247 opened Jul 14, 2025 by yongchaoding Loading…
5 tasks done
5
Add the keyword "template" to member template specialization appears after . or -> in a post-fix expression which is a requirement in C++ standard
#1246 opened Jul 14, 2025 by tomflinda Loading…
9
Mnnvl memory with custom communicator
#1245 opened Jul 14, 2025 by wenscarl • Draft
5 tasks
2
feat: expose python APIs for cutlass blackwell fmha fp8 kernels
#1238 opened Jul 10, 2025 by yzh119 • Draft
5 tasks
7
feat: Restore convenience FLASHINFER_ENABLE_AOT option
#1235 opened Jul 8, 2025 by mgorny Loading…
3 of 5 tasks
7
ProTip! Add no:assignee to see everything that’s not assigned.