flashinfer-ai/flashinfer

Next fix for gemm-allreduce two-shot

Open

#2,359 opened on Jan 15, 2026

View on GitHub
 (10 comments) (0 reactions) (1 assignee)Python (1,031 forks)github user discovery
cute-dslgood first issueop: commop: gemmrefactor

Repository metrics

Stars
 (5,756 stars)
PR merge metrics
 (Avg merge 11d 23h) (186 merged PRs in 30d)

Description

Previously: https://github.com/flashinfer-ai/flashinfer/pull/2171 https://github.com/NVIDIA/cutlass/issues/2845

Next: Update according to https://github.com/NVIDIA/cutlass/blob/main/examples/python/CuTeDSL/distributed/distributed_gemm_all_reduce_blackwell.py

Contributor guide

Research direction
Study the CUTLASS distributed gemm all reduce example and adapt it to the flashinfer codebase, following the patterns from previous PRs #2171 and CUTLASS issue #2845.
Tech stack
pythoncpp
Domain
aibackendperformance
Issue type
Bug
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
4
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-2 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Mostly clear
Prerequisites
CUDAC++Python
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
15