Tested: Fable High Tackles Complex Cross-Border Options Problem

RylanSchaeffer · x · 2026-07-08

A researcher assigned Anthropic's Fable High a complex task: designing a financing structure for the early exercise of startup options across two international jurisdictions. Results showed it performed roughly on par with Gemini 3.1 Pro and GPT-5.5 High, perhaps slightly worse. With a sample size of 1 (N=1), the outcome is somewhat surprising given the prior hype.

Related event: Anthropic's Fable Model Struggles with Complex Financial Structuring(2 posts)→

Original post →

More from Models

Models channel →