Claude Fable 5.1 (Max): benchmark results
Provider: Anthropic. Released 2026-09-01. Access: API.
Aggregate Rating 1958 ± 10, rank #9 of 2168 rated models, from 159 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| ApprenticeBench - API (MCP Tools) | 70 | Cumulative Success Rate (%) | 100 |
| ApprenticeBench - GUI (Computer Use) | 72 | Cumulative Success Rate (%) | 100 |
| Chatbot Arena (Text - Japanese) | 1534 | Arena Score | 100 |
| Chatbot Arena (Text - Korean) | 1514 | Arena Score | 100 |
| EnterpriseBench (CoreCraft Agents) | 77.4 | Score (%) | 100 |
| Humanity's Last Exam No Tools (Mythos 5.1 System Card) | 60.9 | Accuracy (%) | 100 |
| LMArena WebDev Arena | 1765.37 | Arena rating (imported) | 100 |
| Meta Muse Wonder Scorecard - OSWorld V2.1 (mean partial reward) | 80.7 | Mean per-task partial reward (%) | 100 |
| Meta Muse Wonder Scorecard - OSWorld V2.1 (strict binary completion) | 43.1 | Strict task completion (%) | 100 |
| OSWorld 2.0 Partial Score (Opus 5.5 System Card) | 80.7 | Partial Score (%) | 100 |
| OSWorld 2.0 Strict Pass Rate (Mythos 5.1 System Card) | 41.7 | Strict Pass Rate (%) | 100 |
| Vals AI LiveCodeBench | 90.52 | Accuracy (%) | 100 |
Interactive version: theaggregate.ai/model?slug=claude-fable-5-1-max · How It Works · Data refreshed daily, snapshot 2026-10-20.