Maintainer Mode
Sign in with GitHub
Work in progress. Maintainer Mode changes every day: pages, numbers and links can change or break. Tell us what is wrong or missing.

python3Packages.flashinfer-python

Sign in to subscribe
Health
Risk
risk 0 ok
Version
0.7.1 no newer release known to repology
Channels
master 0.7.1 · nixos-unstable 0.6.4 · nixos-26.05 absent master is ahead of nixos-unstable
Advisories
none open
Packaging
meta complete
Description
Library and kernel generator for Large Language Models
Long description
FlashInfer is a library and kernel generator for Large Language Models that provides high-performance implementation of LLM GPU kernels such as FlashAttention, PageAttention and LoRA. FlashInfer focus on LLM serving and inference, and delivers state-of-the-art performance across diverse scenarios.
License
Platforms
31 platforms: Linux, Darwin
Main program
flashinfer
Source provenance
not set: built from source
Outputs
out dist
Status
not broken, not insecure, free
Version bumps 1all open bumps
python3Packages.flashinfer-python: 0.7.1 -> 0.7.2
#574635 by GaetanLepage · opened 1d ago · checks pass
Pull requests 2
python3Packages.flashinfer-python: 0.7.1 -> 0.7.2
#574635 by GaetanLepage · opened 1d ago · checks pass
python3Packages.vllm: fix CUDA build draft
#572573 by amarshall · opened 6d ago
Closed or merged in the synced window 2
Issues 0
None open.