41554 blogs · 4,417,066 posts Claim your blog

Record · corpus.blog/posts/01a0879e-aebc-7209-be67-74f1927e727a

Various approaches to parallelizing Muon

main-horse.github.io · published 22 February 2025

A record is what was published and who pointed at it. The text of the post is not held here: read it at the source, or at its Wayback capture.

Read the post at main-horse.github.io Wayback capture

0blogs citing

First cited: —
Most recent: —
Rank this month: not ranked
Links from this post: 8
Words captured: 3,354

Cited by

No blog we hold has cited this post yet.

Links from this post

Link

Host

https://x.com/samsja19/status/1893507503473971664

x.com

https://github.com/MoonshotAI/Moonlight/blob/master/Moonlight.pdf

github.com

https://github.com/ethansmith2000/fsdp_optimizers/blob/main/fsdp_optimizers/muon.py

github.com

https://github.com/ethansmith2000/MuonVariants/blob/new_muon_shard/muon_fsdp2.py

github.com

https://arxiv.org/html/2411.13055

arxiv.org

MI300X vs H100 vs H200 Benchmark Part 1: Training – CUDA Moat Still Alive Read at the source ↗

semianalysis.com

https://github.com/NVIDIA/nccl-tests/issues/212

github.com

https://jax-ml.github.io/scaling-book/roofline/

jax-ml.github.io