Unverified claim: 'GPT-6 Astra' solved 49 of 50 unpublished theoretical physics problems at $3 each
Dr_Singularity · x · 2026-10-15
Account @DrSingularity shared a paper link claiming a model called "GPT-6 Astra" autonomously solved 49 of 50 unpublished theoretical physics research problems at 98% accuracy, for under $3 per problem — and that researchers initially marked several answers wrong before discovering the AI had caught errors in their own calculations.
Caveat: this comes from a third-party account with no confirmation from OpenAI or named researchers; the "GPT-6" naming has no official basis. Treat as unverified.
More from Models
- Skeptic doubts rumored OpenAI model, citing inability to serve Astra reliably — CtrlAltDwayne · 2026-10-25
- AsciiBench: blind voting arena tests which AI draws better ASCII art across 13 models — JasonBotterill · 2026-10-25
- Satirical ARC-AGI-3 leaderboard shows GPT-6 near 90% score at $10K+ per task — haider1 · 2026-10-25
- Rumor: Anthropic to drop Opus 5.5 two weeks after calling for frontier pacing — StewartalsopIII · 2026-10-25
- Anthropic appears to be quietly nerfing Opus 5.5 usage limits, users report — CtrlAltDwayne · 2026-10-25
- Codex user reports usage down 31% after reset, calls OpenAI pricing uneconomical — CtrlAltDwayne · 2026-10-25