Revision History
Disaggregated serving · 6 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.
- Expanded attention-FFN disaggregation into a full section linking the new hub article, added a 2026 heterogeneous hardware deployments section (AWS Trainium + Cerebras CS-3, AMD Helios + Cerebras WSE, NVIDIA Vera Rubin NVL72 + Groq 3 LPX with its three operating modes), summarized the three 2026 AFD analysis papers, fixed the StepMesh citation, and appended references 24-34.
Version 6 · Sep 7, 2026, 03:23 AM
- Added 6 contextual internal links
Version 4 · Jul 23, 2026, 03:30 PM