Updated daily at 05:00 UTC
Benchmark
Earthborne Rangers (EBR-bench)
Measures: Whether a model improves with practice: it plays the obscure card-based board game Earthborne Rangers ten times, taking notes, and is scored on the objectives it reaches (of 21) in its better scored playthrough.
- Maintainer
- Epoch AI
- Introduced
- 2026
- Status
- Active
- Saturation
- Headroom
- Typical provenance
- Independent
- In scoring set
- Yes
Leaderboard
Every result we hold, whoever ran it. Only independent and third-party rows count towards a score.
| Model | Objectives reached | Provenance | Source | Evaluated |
|---|---|---|---|---|
| GPT-6 Astra OpenAI · Max effort | 76.2% | Independent | Epoch AI | 30 Aug 2026 |
| Claude Opus 5.5 Anthropic · Max effort | 71.4% | Independent | Epoch AI | 22 Sep 2026 |
| Claude Fable 5.1 Anthropic · Max effort | 57.1% | Independent | Epoch AI | 8 Sep 2026 |
| GPT-6 Sol OpenAI · Max effort | 53.3% | Independent | Epoch AI | 22 Sep 2026 |
| Claude Opus 5 Anthropic · Max effort | 45.7% | Independent | Epoch AI | 5 Sep 2026 |
| GPT-5.6 Sol OpenAI · Max effort | 44.8% | Independent | Epoch AI | 4 Sep 2026 |
| Claude Fable 5 Anthropic · Max effort | 39.5% | Independent | Epoch AI | 17 Jul 2026 |
| GPT-5.5 OpenAI · Xhigh effort | 34.3% | Independent | Epoch AI | 27 Jul 2026 |
| Grok 4.6 SpaceXAI (xAI) · Xhigh effort | 30.5% | Independent | Epoch AI | 18 Aug 2026 |
| Claude Opus 4.8 Anthropic · Max effort | 28.6% | Independent | Epoch AI | 7 Aug 2026 |
| GPT-5.4 OpenAI · Xhigh effort | 25.4% | Independent | Epoch AI | 25 Jun 2026 |
| GPT-5.2 OpenAI · Xhigh effort | 23.0% | Independent | Epoch AI | 26 Jun 2026 |
| Claude Opus 4.7 Anthropic · Max effort | 19.0% | Independent | Epoch AI | 30 Jun 2026 |
| Claude Opus 4.5 Anthropic · 128K effort | 14.3% | Independent | Epoch AI | 25 Jun 2026 |
| Gemini 3.1 Pro | 14.3% | Independent | Epoch AI | 25 Jun 2026 |
| Claude Opus 4.6 Anthropic · Max effort | 12.7% | Independent | Epoch AI | 29 Jun 2026 |
| GPT-5 OpenAI · High effort | 12.7% | Independent | Epoch AI | 29 Jun 2026 |
| GLM-5.2 Z.ai (Zhipu AI) · Max effort | 9.5% | Independent | Epoch AI | 29 Jun 2026 |
| Qwen3.7-Max | 9.5% | Independent | Epoch AI | 25 Jun 2026 |
| Claude Opus 4.1 | 7.9% | Independent | Epoch AI | 25 Jun 2026 |
| Gemini 3.5 Flash Google DeepMind · High effort | 4.8% | Independent | Epoch AI | 25 Jun 2026 |
| Claude Sonnet 4.5 | 2.4% | Independent | Epoch AI | 25 Jun 2026 |
| Kimi K2.6 | 2.4% | Independent | Epoch AI | 25 Jun 2026 |
- Epoch AI, 'Capabilities & benchmarking', epoch.ai/benchmarks, CC BY 4.0.