GPT-5.6 Pro Solves 5 Erdős Problems
ThunderBeanage · reddit · 2026-07-14
The author claims that during a week of stealth testing, they used GPT-5.6 Pro to solve multiple Erdős problems. They have already submitted five of them—730, 671, 948, 346, and 1139—to the site, with some still pending verification.
They also mentioned previously working on solving 728 (reportedly the first Erdős problem solved by AI) and using 5.4 Pro to solve 1196, which later led to a co-authored paper with Jared Lichtman, Terence Tao, and others based on that method. The author's main takeaway is that Erdős problems are actually solved much more frequently than social media suggests; many just don't get widely publicized. They make it a habit to re-run these problems every time a new model is released.
Related event: GPT-5.6 reportedly cracks decades-old Erdős problems amid hype debate(10 posts)→
More from Models
- Rumor: GPT-6 obsolete for months as OpenAI teases math breakthroughs from new frontier model — ChrisGPT · 2026-10-27
- Meta's Muse Spark 1.3 System Card Relies on Four Tübingen Benchmarks — maksym_andr · 2026-10-27
- ChatGPT Pro 500 Spotted: £445/Month Tier With Shared Weekly Usage Limits — yoimnotkesku · 2026-10-27
- Experts Clash: Today's Models Dwarf GPT-4, But Journal Delays Lock Out New LLMs — DeryaTR_ · 2026-10-27
- OpenAI's Terra release disappoints: community senses a widening lab-to-public frontier gap — ChrisGPT · 2026-10-27
- Claude spots broken physics in AI videos, fixes them with CFD and finite-element verification — OfirPress · 2026-10-27