GPT-6 Astra tops RSI-Exam at 0.5126, 18.4% above GPT-5.6 Sol, unverified claim says
HuaxiuYaoML · x · 2026-09-09
A third-party account claims OpenAI's GPT-6 Astra scored 0.5126 on the self-styled RSI-Exam benchmark, taking #1 — 18.4% above GPT-5.6 Sol and 54.8% above GPT-5.5, calling it a generational leap for recursive self-improvement.
Caveat: the claim and the benchmark are not officially confirmed by OpenAI; treat with skepticism.
Related event: GPT-6 Astra Tops RSI-Exam Benchmark, Up 18.4% Over GPT-5.6(2 posts)→
More from Models
- Decart CEO teases Solaris, an interface world model — c_valenzuelab · 2026-09-09
- GPT-6 Astra review: brilliant at 3D games and strategy, but unreliable instruction-following blocks daily use — petergyang · 2026-09-09
- Open-source 4B VLM RL-trained to play GeoGuessr beats GPT 5.4 mini and Haiku on one A100 — SergioPaniego · 2026-09-09
- Sam Altman says unprecedented Astra demand may force OpenAI to pause new Pro subscriptions — sama · 2026-09-09
- GPT-6 Astra tops computer-use benchmarks but trails Fable 5.1 — why first impressions are so split — The AI Daily Brief · 2026-09-09
- User finds GPT-6 Astra weak at orchestration, would use it as expert call-in only — koltregaskes · 2026-09-09