Blueprint-Bench 2 (Fable/Mythos): leaderboard
Spatial-reasoning benchmark measuring how accurately models convert apartment photos into 2D floor plans. This slice reports Anthropic's own runs from the Claude Fable 5 / Mythos 5 system card; small frontier-model pool.
Metric: Score (0-1). Source: www-cdn.anthropic.com. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Fable 5 | 0.39 | #16 |
| 2 | Claude Mythos 5 | 0.39 | #2 |
| 3 | GPT-5.5 | 0.36 | #31 |
| 4 | Gemini 3.5 Flash | 0.34 | #60 |
| 5 | Gemini 3.1 Pro | 0.27 | |
| 6 | Claude Opus 4.8 | 0.14 | #37 |
Interactive version: theaggregate.ai/benchmark?slug=blueprint-bench-2-fable-mythos · How It Works · Data refreshed daily, snapshot 2026-10-20.