All benchmarksIndependent evaluation · Epoch AI
SpireClimb
Each model’s result as Epoch AI’s Benchmarking Hub publishes it, run by Epoch or collected from the source it cites. Facthory has not run them.
Data: Epoch AI, ‘Capabilities & benchmarking’, epoch.ai/benchmarks, licensed CC BY 4.0. Read on 16 Nov 2026.
| # | Model | Organization | Result | Evidence |
|---|---|---|---|---|
| 1 | GPT-6 Astra (max) API access | OpenAI | 34% ± 4.1% | – |
| 2 | GPT-6.1 Sol (max) API access | OpenAI | 23.8% ± 3.7% | – |
| 3 | GPT-6.1 Terra (max) API access | OpenAI | 22.3% ± 3.4% | – |
| 4 | Claude Fable 5.2 (max) API access | Anthropic | 15.3% ± 3.1% | – |
| 5 | GPT-6 Sol (max) API access | OpenAI | 9.2% ± 2.4% | – |
| 6 | Claude Opus 5.5 (max) API access | Anthropic | 5.8% ± 2% | – |
| 7 | Claude Fable 5.1 (max) API access | Anthropic | 2% ± 1.1% | – |
| 8 | Claude Sonnet 5.5 (max) API access | Anthropic | 1.3% ± 0.9% | – |
| 9 | GPT-6 Luna (max) API access | OpenAI | 0% ± 0% | – |
