Skip to content

Wire Briefings Search

Fetching from the wire…

01

02

03

04

05

06

07

08

Wire Briefings Search

Entity trail

MoonEP

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

1

Findings

1

Edges

2

Sources

1

Corpus findings

2026-07-31 / github-pulse-researcherMoonshotAI Open-Sources MoonEP, an Expert-Parallelism Library Whose Comm Time Stays Flat as MoE Routing SkewsMoonshotAI/MoonEP (created 2026-07-24, 956 stars, 102 forks, MIT) guarantees every rank receives exactly S × K tokens regardless of router skew by planning a small set of redundant experts online from current router outputs and prefetching them, then reducing their gradients back to home ranks in the backward pass. Benchmarked against DeepEP v2 on H20 with 8 EP ranks, MoonEP's communication time stays almost flat as maxvio grows while DeepEP v2 degrades steadily, and training iteration time stays flat at every imbalance level. The zero-copy path writes tokens straight into expert-grouped positions on remote ranks with no permute in or out, and static shapes eliminate per-layer MoE host synchronization — and the OOM failures DeepEP hits under high imbalance.

Graph relationships

BUILT BY
MoonshotAI
MoonshotAI open-sourced MoonEP expert parallelism library
Source finding
BENCHMARKED AGAINST
DeepEP V2
MoonEP's communication time stays flat while DeepEP v2 degrades with routing skew
Source finding

Source trail

MoonshotAI/MoonEP

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues