Epoch AI: GPT-6 Astra tops frontier models in Slay the Spire 2 but still loses to top humans
sandersted · x · 2026-10-15
Epoch AI tested frontier models on Slay the Spire 2 at max difficulty (Ascension 10): 15 runs per character, 75 runs per model, 90% confidence intervals.
- Leading model GPT-6 Astra wins 32%, GPT-6.1 Sol is similar, GPT-6 Sol 12%
- Claude Opus 5.5 5%; Fable 5.1 and Sonnet 5.5 2-3%; GPT-6 Luna went 0/75
- The sequel is widely considered easier and win rates are higher than the first game, yet all models remain far below estimated top-human win rates
More from Models
- Silence Is Billable on Voice Models: Even Dead Air Costs Money — sophiamyang · 2026-10-15
- NCM: new text generation paradigm up to 31x faster than diffusion LMs with better quality — simonguozirui · 2026-10-15
- Gemma 4 26B priced at $0.07/M input tokens, ~40% below comparable providers — testingcatalog · 2026-10-15
- Google execs cite early signs of recursive self-improvement behind 3-4 week Gemini cadence — danielrock · 2026-10-15
- MoralityBench tests 16 LLMs with validated moral psychology surveys; only 3 exceed 90% consistency — NakedPlato · 2026-10-15
- Claude systematically debunks the speculative EDAR-selenium hypothesis about East Asians — teortaxesTex · 2026-10-15