GPT-6-Astra solves 49 of 50 physics research problems at $2.40 apiece
The Decoder · rss · 2026-10-15
On the new TheoPhysEval benchmark, GPT-6-Astra solved 49 of 50 unpublished theoretical physics calculations at expert-level accuracy, at about $2.40 per problem — and in some apparent failures the experts' reference solutions turned out to be wrong. Researchers note physics hasn't seen math-like breakthroughs because agents lack "taste," but still expect autonomous top-level AI research soon, with implications for how young physicists are trained.
More from Models
- User math: ~700M tokens daily quota forces mid-afternoon resets, 28-day event seen as a stopgap — yangyi · 2026-10-23
- Rumor: Anthropic training a model matching OpenAI's internal one; big launches in December — imjustnewatai · 2026-10-23
- AI researcher: Claude finally finished tedious research tasks I gave up on years ago — MikePFrank · 2026-10-23
- Dream Labs emerges from stealth with Dream-1 robot foundation model at $3/hour — scott_e_reed · 2026-10-23
- Claude Models on Google Cloud (Anthropic managed) is now generally available — claudeai · 2026-10-23
- Beam vs GLM 5.3 Abliterated: 10-round attack/defense test shows 96% of vulns patched — bigblueboo · 2026-10-23