All benchmarksIndependent evaluation · Epoch AI
GDPval
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-5.2 (none) API access | OpenAI | 49.7% | – |
| 2 | Claude Opus 4.5 (no thinking) API access | Anthropic | 45.5% | – |
| 3 | claude-opus-4-1-20250805 API access | Anthropic | 43.6% | – |
| 4 | Claude Sonnet 4.5 (no thinking) API access | Anthropic | 42.5% | – |
| 5 | Gemini 3 Pro Preview API access | Google DeepMind | 40.3% | – |
| 6 | GPT-5 (medium) API access | OpenAI | 34.8% | – |
| 7 | o3 (medium) API access | OpenAI | 30.8% | – |
| 8 | o4-mini (high) API access | OpenAI | 25.3% | – |
| 9 | gemini-2.5-pro API access | Google DeepMind | 23.3% | – |
| 10 | grok-4-0709_high API access | xAI | 21.1% | – |
| 11 | GPT-4o (Nov 2024) API access | OpenAI | 9.9% | – |
