Artificial Analysis (@ArtificialAnlys) · 10/23/2026, 02:12:51
CritPt v2 Launches: Frontier Models Surge on Research-Level Physics Benchmark
Artificial Analysis has launched CritPt v2, a revised research-level physics benchmark that supports code execution. Recent models like GPT-6 Astra top the leaderboard with 83% accuracy, marking a 54-point improvement over GPT-5 from August 2025 and highlighting rapid progress in scientific reasoning.
SOURCE COVERAGEFull Original Report
Artificial Analysis (@ArtificialAnlys) · 2026-10-22 18
CritPt v2 has launched on Artificial Analysis: the CritPt authors have reviewed all 70 research-level physics challenges, and models now solve them with a code-execution tool
Following feedback from the research community, the CritPt team and physics expert collaborators reviewed every challenge, corrected underspecified statements, errors, and typos, and improved the stability of the grader. We now run CritPt v2 agentically: models can write and run code in an offline sandbox with tools for computation.
We’ll be updating the Intelligence Index in late October, including changes to CritPt.
Key takeaways:
➤ Recent OpenAI releases top the leaderboard: GPT-6 Astra (max) and GPT-6.1 Sol (max) score highest at 83% and 82%, followed by Claude Opus 5.5 (max with fallback) at 78% and GPT-5.6 Sol (max) at 75%
➤ CritPt v2 shows the rapid progress at the frontier on research-level physics: GPT-5 (high), released in August 2025, scores 29% on CritPt v2, 54 points below GPT-6 Astra (max)
Thank you to @MiniHui_zhu, @MinyangTian1, @haopeng_uiuc and the broader CritPt team for their work on this revision and their continued collaboration
