FlashMLA Files

FlashMLA: Efficient Multi-head Latent Attention Kernels

This is an exact mirror of the FlashMLA project, hosted at https://github.com/deepseek-ai/FlashMLA. SourceForge is not affiliated with FlashMLA.

The interactive file manager requires Javascript. Please enable it or use sftp or scp.
You may still browse the files here.

Name	Modified	Size
FlashMLA source code.tar.gz	2026-09-30	299.3 kB
FlashMLA source code.zip	2026-09-30	399.7 kB
Totals: 2 Items		699.1 kB

Other Useful Business Software

Save Up to 91% on Cloud Compute With Spot VMs

Automatic sustained-use discounts. One free VM per month. No negotiation needed.

Run batch jobs at 60-91% off with Spot VMs. Long-running workloads get automatic discounts with sustained use.

Start Free

Earn up to 16% annual interest with Nexo.

Let your crypto work for you

Put idle assets to work with competitive interest rates, borrow without selling, and trade with precision. All in one platform. Geographic restrictions, eligibility, and terms apply.

Get started with Nexo.

Recommended Projects

Kimi Linear
An expressive, efficient attention architecture
Transformer Engine
A library for accelerating Transformer models on NVIDIA GPUs
FlashAttention
Fast and memory-efficient exact attention
1m-go-tcp-server
Benchmarks for implementation of servers
MLA Auto-Generator
Generates correctly formatted(MLA, APA, Chicago, CPE, Turabian) entries for works cited or bibliography pages.