On the Chartography benchmark (100 questions over dense real-world charts), Fabl
Source
ClaudeDevs
Author
ClaudeDevs
Date
Why it matters
If you're building agents that read dense charts or documents, giving the model a zoom/crop tool can more than double accuracy — the Chartography numbers show tool-augmented perception beats raw model vision, with a cookbook to replicate it.
On the Chartography benchmark (100 questions over dense real-world charts), Fable 5's accuracy goes from 29% to 73% with a zoom tool, and Sonnet 5’s from 13% to 44%.
https://t.co/ic4aW4z9Iq
Transcript
On the Chartography benchmark (100 questions over dense real-world charts), Fable 5's accuracy goes from 29% to 73% with a zoom tool, and Sonnet 5’s from 13% to 44%.
See the cookbook to learn more:
https://t.co/ic4aW4z9Iq
