Chartography Benchmark: Zoom Tool Boosts Claude's Chart Accuracy to 44%
echen · x · 2026-07-28
SurgeAI introduced the Chartography benchmark, designed to evaluate how well models read and interpret professional charts.
In a test of 100 questions based on dense real-world charts, both Fable 5 and Claude 3.5 Sonnet showed significant improvement when equipped with a zoom tool. Fable 5's accuracy jumped from 29% to 73%, while Sonnet's increased from 13% to 44%.
Related event: Claude's Zoom Tool Significantly Boosts Chart Recognition Accuracy(3 posts)→
More from Models
- Kimi K3 beats GLM-5.2 in exploit tests but still fails end-to-end attacks — kevinsxu · 2026-07-28
- Big models are “wiser” while more reasoning makes them more “diligent” — breath_mirror · 2026-07-28
- Microsoft Launches Homegrown AI Security Model, Beating GPT at Half the Cost — MichaelFNunez · 2026-07-28
- Testing K3 and Open Models: Reasoning Tokens Can Burn Entire Budgets — MaziyarPanahi · 2026-07-28
- Local Test: Nanbeige4.2-3B Lags Behind Qwen MoE in KV Cache Efficiency — TechTefa · 2026-07-28
- 35B Agentic Bakeoff: KAT-Coder Matches Qwen at Half the Token Cost — IvGranite · 2026-07-28