Revision History
CUDA · 13 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: fixes from an independent fact-check pass (see /corrections)
Version 13 · Sep 15, 2026, 11:33 PM
- CUDA Toolkit 13.4 section: final copy pass
Version 12 · Sep 15, 2026, 10:42 PM
- Add CUDA Toolkit 13.4 coverage (Windows on Arm, Rubin sm_107 preview, MPS V3, Compute Fabric Transport, CDMM default, CUDA Python and CCCL 3.4)
Version 11 · Sep 15, 2026, 10:40 PM
- Correction: replaced outdated release, universal AI and speedup claims, and conflated hardware material with a primary-source account of CUDA's model, memory, toolchain, libraries, compatibility, licensing, and portability. Correct the PyTorch CUDA Semantics source-update date to the official July 17, 2026 record.
Version 10 · Jul 29, 2026, 06:04 AM
- Added 6 contextual internal links
Version 9 · Jul 23, 2026, 03:00 PM