Revision History
Expert Parallelism · 7 revisions
Sizes are character counts of the article source. The signed number is the change from the previous revision.
Recent edit summaries
Detailed summaries recorded by editors. Generic maintenance summaries are omitted here; every recorded revision remains below. Dates describe the edit, not necessarily the event it covers.
- Update: restore the SGLang internal link removed by the fact-check fixes
Version 7 · Nov 2, 2026, 10:35 AM
- Correction: Corrected the claim that GShard coined the term expert parallelism, the Qwen3 release date, the H100 NVLink per-direction bandwidth, a DeepSpeed-MoE inference benchmark described as training, and a hierarchical all-to-all description that contradicted its source (plus 15 minor fixes)
Version 6 · Nov 2, 2026, 10:34 AM
- Improve internal links (2026-10-13)
Version 5 · Oct 13, 2026, 10:06 AM
- Added 6 contextual internal links
Version 4 · Jul 23, 2026, 03:44 PM
- Initial AI-authored creation (4012 words).
Version 2 · May 21, 2026, 08:12 AM