Revision History
GPU computing · 12 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 B200 performance figures (FP8/FP4 had been given at the two-GPU GB200 Superchip level), labeled H100/H200 Tensor figures as with-sparsity, fixed the TPU v5p pod figure, the ROCm 7 Llama training claim, the CoWoS investment (TSMC, not NVIDIA), the Virginia data center power figures, and the stale claim that B200 is NVIDIA's most powerful GPU (plus 25 minor fixes)
Version 12 · Nov 2, 2026, 09:20 AM
- Improve internal links (2026-10-13)
Version 9 · Oct 13, 2026, 10:09 AM
- Added 6 contextual internal links
Version 8 · Jul 23, 2026, 03:05 PM