# EBR-bench leaderboard

external benchmark

> Every figure below is reproduced as its upstream source published it: nothing is modelled, estimated, interpolated or converted. `Unavailable` means no source published the value — it is never a zero. Each value carries its evidence state and the date it was observed.

Page: https://modelchoice.dev/leaderboards/benchmarks/external-ebr-bench

Earthborne Rangers (EBR-bench). Every model the catalog carries a published EBR-bench value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | external |
| Measure | Long-horizon game play with notes across playthroughs |
| Tasks | Ten playthroughs of an Earthborne Rangers campaign segment per sample (eight learning, two scored) |
| Difficulty | Learning an unfamiliar game over multi-hour playthroughs |
| Published by | [Earthborne Rangers (EBR-bench), Epoch AI](https://epoch.ai/benchmarks/ebr-bench) |

Epoch AI's EBR-bench tests whether models get better at an unfamiliar game with practice: they play the campaign card game Earthborne Rangers repeatedly, take notes, and are scored on later playthroughs.

## EBR-bench ranking

33 models with a published EBR-bench value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.

| Rank | Model | Provider | Long-horizon game play with notes across playthroughs | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [GPT-6.2 Astra](https://modelchoice.dev/models/gpt-6-2-astra) | OpenAI | 92.4 | Observed 2026-11-21 · source benchlm:models |
| 2 | [Claude Fable 5.2](https://modelchoice.dev/models/claude-fable-5-2) | Anthropic | 90.5 | Observed 2026-11-21 · source benchlm:models |
| 3 | [Gemini 4 Carbon](https://modelchoice.dev/models/gemini-4-carbon) | Google | 87.6 | Observed 2026-11-21 · source benchlm:models |
| 4 | [GPT-6.1 Terra](https://modelchoice.dev/models/gpt-6-1-terra) | OpenAI | 78.1 | Observed 2026-11-21 · source benchlm:models |
| 5 | [GPT-6 Astra](https://modelchoice.dev/models/gpt-6-astra) | OpenAI | 76.2 | Observed 2026-11-21 · source benchlm:models |
| 6 | [Claude Opus 5.5](https://modelchoice.dev/models/claude-opus-5-5) | Anthropic | 71.4 | Observed 2026-11-21 · source benchlm:models |
| 7 | [Gemini 4 Hydrogen](https://modelchoice.dev/models/gemini-4-hydrogen) | Google | 70.5 | Observed 2026-11-21 · source benchlm:models |
| 8 | [Gemini 4 Argon](https://modelchoice.dev/models/gemini-4-argon) | Google | 64.8 | Observed 2026-11-21 · source benchlm:models |
| 9 | [Claude Fable 5.1](https://modelchoice.dev/models/claude-fable-5-1) | Anthropic | 57.1 | Observed 2026-11-21 · source benchlm:models |
| 10 | [GPT-6.1 Sol](https://modelchoice.dev/models/gpt-6-1-sol) | OpenAI | 54.3 | Observed 2026-11-21 · source benchlm:models |
| 11 | [GPT-6 Sol](https://modelchoice.dev/models/gpt-6-sol) | OpenAI | 53.3 | Observed 2026-11-21 · source benchlm:models |
| 12 | [Claude Opus 5](https://modelchoice.dev/models/claude-opus-5) | Anthropic | 45.7 | Observed 2026-11-21 · source benchlm:models |
| 13 | [GPT-5.6 Sol](https://modelchoice.dev/models/gpt-5-6-sol) | OpenAI | 44.8 | Observed 2026-11-21 · source benchlm:models |
| 14 | [Claude Haiku 5.5](https://modelchoice.dev/models/claude-haiku-5-5) | Anthropic | 43.8 | Observed 2026-11-21 · source benchlm:models |
| 15 | [Claude Fable 5](https://modelchoice.dev/models/claude-fable-5) | Anthropic | 39.5 | Observed 2026-11-21 · source benchlm:models |
| 16 | [Kimi K3.1 Pro](https://modelchoice.dev/models/kimi-3-1-pro) | Moonshot AI | 37.1 | Observed 2026-11-21 · source benchlm:models |
| 17 | [GPT-5.5](https://modelchoice.dev/models/gpt-5-5) | OpenAI | 34.3 | Observed 2026-11-21 · source benchlm:models |
| 18 | [Muse Wonder](https://modelchoice.dev/models/muse-wonder) | Meta | 31.4 | Observed 2026-11-21 · source benchlm:models |
| 19 | [Grok 4.6](https://modelchoice.dev/models/grok-4-6) | xAI | 30.5 | Observed 2026-11-21 · source benchlm:models |
| 20 | [Claude Opus 4.8](https://modelchoice.dev/models/claude-opus-4-8) | Anthropic | 28.6 | Observed 2026-11-21 · source benchlm:models |
| 21 | [GPT-5.4](https://modelchoice.dev/models/gpt-5-4) | OpenAI | 25.4 | Observed 2026-11-21 · source benchlm:models |
| 22 | [GPT-5.2](https://modelchoice.dev/models/gpt-5-2) | OpenAI | 23.0 | Observed 2026-11-21 · source benchlm:models |
| 23 | [Claude Opus 4.7](https://modelchoice.dev/models/claude-opus-4-7) | Anthropic | 19.0 | Observed 2026-11-21 · source benchlm:models |
| 24 | [Claude Opus 4.5](https://modelchoice.dev/models/claude-opus-4-5) | Anthropic | 14.3 | Observed 2026-11-21 · source benchlm:models |
| 24 | [Gemini 3.1 Pro](https://modelchoice.dev/models/gemini-3-1-pro) | Google | 14.3 | Observed 2026-11-21 · source benchlm:models |
| 26 | [Claude Opus 4.6](https://modelchoice.dev/models/claude-opus-4-6) | Anthropic | 12.7 | Observed 2026-11-21 · source benchlm:models |
| 26 | [GPT-5](https://modelchoice.dev/models/gpt-5) | OpenAI | 12.7 | Observed 2026-11-21 · source benchlm:models |
| 26 | [GPT-5 (high)](https://modelchoice.dev/models/gpt-5-high) | OpenAI | 12.7 | Last good 2026-11-18 · source benchlm:models |
| 29 | [GLM-5.2](https://modelchoice.dev/models/glm-5-2) | Z.AI | 9.5 | Observed 2026-11-21 · source benchlm:models |
| 29 | [Qwen3.7 Max](https://modelchoice.dev/models/qwen3-7-max) | Alibaba | 9.5 | Observed 2026-11-21 · source benchlm:models |
| 31 | [Gemini 3.5 Flash](https://modelchoice.dev/models/gemini-3-5-flash) | Google | 4.8 | Observed 2026-11-21 · source benchlm:models |
| 32 | [Claude Sonnet 4.5](https://modelchoice.dev/models/claude-sonnet-4-5) | Anthropic | 2.4 | Observed 2026-11-21 · source benchlm:models |
| 32 | [Kimi K2.6](https://modelchoice.dev/models/kimi-2-6) | Moonshot AI | 2.4 | Observed 2026-11-21 · source benchlm:models |

Rows are ordered by the value the source published, highest first. The source does not state whether a higher value is the better result, so this document does not either: for a benchmark that measures a rate of failure, read the table from the bottom.
