Benchmarks · agentic
SpireClimb (Epoch AI)
Epoch AI's Slay the Spire agent benchmark: climb the spire by playing cards and choosing routes. Score is progress up the spire. Data: Epoch AI, CC BY.
7 models · 7 results · data as of 2026-10-29 · epoch.ai ↗
Leaderboard
GPT-6 Astra max
All results
7 rows| # | Model | Effort | SpireClimb (Epoch AI) | Harness |
|---|---|---|---|---|
| 1 | GPT-6 Astra | max | 25.7% | Epoch AI |
| 2 | GPT-6.1 Sol | max | 24.7% | Epoch AI |
| 3 | GPT-6 Sol | max | 7.5% | Epoch AI |
| 4 | Opus 5.5 | max | 3.3% | Epoch AI |
| 5 | Fable 5.1 | max | 2.0% | Epoch AI |
| 6 | Sonnet 5.5 | max | 1.3% | Epoch AI |
| 7 | GPT-6 Luna | max | 0.0% | Epoch AI |