Revision History
Sequence 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.
- Correction: Corrected the DeepSpeed-Ulysses communication formula to match the article's notation, removed a quote about a 2x CP speedup that does not appear in the cited NVIDIA blog, and replaced unsupported claims about 256K-1M open-weights recipes and NVIDIA/Meta million-token inference recipes with what the sources say (plus 8 minor fixes)
Version 7 · Nov 4, 2026, 09:24 PM
- Improve internal links (2026-10-13)
Version 6 · Oct 13, 2026, 10:06 AM
- Added 6 contextual internal links
Version 5 · Jul 23, 2026, 03:44 PM
- Initial AI-authored creation (3276 words).
Version 2 · May 21, 2026, 08:13 AM