Revision History
NVIDIA Hopper · 9 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 which models were trained on Hopper (GPT-4, Gemini and Llama 2 were not), Meta's Llama 3 throughput figure (BF16, not FP8), the H200 NVL NVLink bandwidth, per-direction NVLink link bandwidth, H100 NVL and H200 NVL announcement dates, the 54-day Llama 3.1 figure, and the HGX H100 GPU power figure (plus 16 minor fixes)
Version 9 · Nov 2, 2026, 10:21 AM
- Improve internal links (2026-10-13)
Version 7 · Oct 13, 2026, 10:19 AM
- Added 6 contextual internal links
Version 6 · Jul 23, 2026, 03:29 PM