AutomationBench (Fable/Mythos): leaderboard
Zapier benchmark for end-to-end business workflow automation across sales, marketing, operations, and support tasks. This slice reports Anthropic's own runs from the Claude Fable 5 / Mythos 5 system card; small frontier-model pool.
Metric: Score (%). Source: www-cdn.anthropic.com. Saturation forecast: Around 2030. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Fable 5 | 17.4 | #16 |
| 2 | Claude Mythos 5 | 17.4 | #2 |
| 3 | Claude Opus 4.8 | 15.5 | #37 |
| 4 | Gemini 3.5 Flash | 14.5 | #60 |
| 5 | GPT-5.5 | 12.9 | #31 |
| 6 | Gemini 3.1 Pro (Preview) | 9.6 | #58 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=automationbench-fable-mythos · How It Works · Data refreshed daily, snapshot 2026-10-20.