No model sweeps AA v5: Opus 5.5 leads three of nine evaluations
ArtificialAnlys · x · 2026-10-28
AA v5 leaves significant headroom across all nine evaluations, with no single model leading them all: Claude Opus 5.5 tops FrontierCode 2.0 (46%), HLE-Diamond (68%), and AA-Omniscience v1.1 (48); GPT-6 Astra leads Terminal-Bench-Science (63%) and CritPt v2 hard subset (72%); Claude Sonnet 5.5 leads AA-Briefcase (1812 Elo); Kimi K3.1 Pro leads Terminal-Bench 4.0 (67%); Gemini 4 Argon leads AutomationBench-AA (78%); Muse Wonder Full leads GDP.pdf (33% all-pass). CritPt v2 uses a private 40-question hard subset run in an offline Python sandbox via the Stirrup agentic harness.
More from Models
- Commenter predicts OpenAI must ship an Opus 5.5 competitor before its 28-day shipping event ends — CtrlAltDwayne · 2026-10-28
- Steering Vector Recipe Turns Qwen 27B Into "Evil Qwen" — basedjensen · 2026-10-28
- MoveBench debuts: a rubric-based benchmark for LLM sports and health coaching — Yitong Li · 2026-10-28
- Debate: Claude isn't 'lobotomized' — inference stack updates may shift old-model outputs — JasonBotterill · 2026-10-28
- AI labs near IPO, yet LLMs still produce 'astonishingly terrible prose' by default — GabGarrett · 2026-10-28
- Rebuttal: Kimi 3.1 Nears Fable on Frontier Code 2, Contradicting 'China Models Crushed' Claim — bookwormengr · 2026-10-28