GPT-6 reportedly solves a SimpleBench question models have failed for years
ChrisGPT · x · 2026-09-05
A user reports that GPT-6, in a fresh chat with no memory or history, finally answered a SimpleBench question that models have been struggling with for years, suggesting a notable jump in SimpleBench score from GPT-5.6. This is an unverified anecdote; official and third-party benchmarks remain to be seen.
Related event: Leak Claims GPT-6 Finally Solves Long-Standing SimpleBench Problem(2 posts)→
More from Models
- Developer says OpenAI silently kicked him out of cyber Trusted Access program after full identity verification — QuixiAI · 2026-09-05
- 'Astra High' tier revealed as dev reports only 3% quota drop — chrisalbon · 2026-09-05
- Switching models only cut this dev's usage quota by 3% — chrisalbon · 2026-09-05
- Gemini 3.8 Flash arrives in GitHub Copilot, strong on terminal coding tasks — mariorod1 · 2026-09-05
- GPT-6 Astra lands in GitHub Copilot with focus on long-horizon autonomous coding — mariorod1 · 2026-09-05
- GPT-6 Astra flunks complex PCB routing after 2h20m and 15% of weekly limits in biggest public test — yacineMTB · 2026-09-05