Epoch AI game benchmark: GPT-6.1 Sol leads Slay the Spire but still trails top humans
Jsevillamol · x · 2026-10-15
Epoch AI's game benchmark tracking shows frontier models improving quickly at games like Slay the Spire, with OpenAI currently leading Anthropic on this benchmark and GPT-6.1 Sol topping the original game's win rates. Still, on the hardest difficulty, all frontier models remain far below top human players.
Jsevillamol notes AI has improved significantly since they started streaming the benchmark.
More from Models
- Model greets evaluators in its chain of thought — 4 trends paint an alarming AI trajectory — dfrsrchtwts · 2026-10-15
- TypeSafe AI ships Jev 1.14.0 with experimental images hidden in danger-zone flags — michellechen · 2026-10-15
- ADL says it works with OpenAI, Anthropic, Google, Meta and xAI to curb antisemitism; Claude translation blocked by safety filter — amasad · 2026-10-15
- Fable 5.5 adoption or API price cuts: one bet on what it takes to reignite growth — zephyr_z9 · 2026-10-15
- GPT-6 Astra tops AA-Robotics v0.1, but frontier LLMs remain ~100x slower than humans at robot arm tasks — ArtificialAnlys · 2026-10-15
- Token prices are subsidized like early Uber, and GPT-6 Astra insisted it was right on physics evals — UltraRareAF · 2026-10-15