Revision History
AIME 2025 · 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.
- Update: regenerate results chart from the fact-checked results table
Version 13 · Nov 2, 2026, 06:40 AM
- Correction: Corrected numerous AIME 2025 scores and labels: GPT-5's with-Python 99.6% shown as closed book, GPT-5 no-thinking, o1 (79.2%), original DeepSeek R1 (70.0%), Claude 3.7 Sonnet (AIME 2024 figure), Grok 4 Heavy and Gemini 3 Pro tool scores in the no-tools table, GLM-5/5.1 and Kimi K2.6 AIME 2026 scores presented as AIME 2025, and removed unsupported or fabricated results for Gemini 3.1 Pro, Claude Opus 4.6 with Python, Kimi K2.5 with Python, DeepSeek V3.2 Speciale with Python, DeepSeek V4 Pro Max, and other errors (plus 21 minor fixes)
Version 12 · Nov 2, 2026, 06:39 AM
- Improve internal links (2026-10-13)
Version 9 · Oct 13, 2026, 10:01 AM
- Added 6 contextual internal links
Version 8 · Jul 23, 2026, 03:00 PM
- Pre-update snapshot before enhancement to 5677 words.
Version 3 · May 18, 2026, 05:24 AM