huggingface / huggingface/candle

Why not use NV's cutlass for better performance, intstead of re-create computations by cu?

Open
#2,763 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
Rust
Stars
21k
Forks
1.8k
Avg merge
16h 42m
Merged PRs (30d)
25

Description

https://github.com/NVIDIA/cutlass/blob/main/LICENSE.txt

Contributor guide

No contributing guide indexed for this repository

Research direction

The issue contains only a link to NVIDIA CUTLASS's LICENSE.txt and names no Candle file, test, or reproduction. Start by reviewing the linked license and locating the Candle computations referred to in the issue; the work needs an agreed integration scope and a defined performance comparison before completion can be assessed.

Written by the indexing model from the issue text.

Assessment

Tech stack
rust
Domain
machine-learning, performance
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
15/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.