Frontier AI models now win 90%+ of Slay the Spire runs, up from barely winning six months ago
NathanpmYoung · x · 2026-10-15
Epoch AI's new SpireClimb benchmark tracks how well frontier models play Slay the Spire. Six months ago they could barely beat the game; now they win over 90% of runs at the easiest difficulty and up to 20% at the hardest.
The chart shows Ascension 0 win rates: GPT-6 Astra and GPT-6.1 Sol approach 98%, Opus 5.5 and GPT-6 Sol sit around 90%+, Sonnet 5.5 about 85%, Fable 5.1 roughly 62%, and GPT-6 Luna about 22% — with top models nearly matching the estimated top-human level. 15 samples per character, 60 per model, 90% confidence intervals.
More from Models
- 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
- Epoch AI: GPT-6 Astra tops frontier models in Slay the Spire 2 but still loses to top humans — sandersted · 2026-10-15
- Google execs cite early signs of recursive self-improvement behind 3-4 week Gemini cadence — danielrock · 2026-10-15
- Claude systematically debunks the speculative EDAR-selenium hypothesis about East Asians — teortaxesTex · 2026-10-15
- Meta's Muse Wonder (Max) lands at #12 in Agent Arena at 2.1x the cost of its predecessor — arena · 2026-10-15