New: Claude Fable 5.2 eval report is here.Read the report ↗

Best LLMs for professional chart understanding

Chartography

Tests reading professional charts accurately, helping you choose a model to support client reports and presentations with less manual interpretation. Results cover 100 tasks; tool-free and tool-assisted settings are recorded separately.

Updated: Nov 9, 2026Data as of: Nov 9, 2026

Limited reportingNot saturated

Official benchmark

Chartography results. Each row identifies its benchmark version and evaluation track. Scores from different versions are not directly comparable. Best effort from the latest stored report date per model and version.
Model / creator	Version	Reported score
Claude Fable 5.2Anthropic · Proprietary	Anthropic implementation / Gemini 3.5 Flash grader	91.6%
Gemini 4 CarbonGoogle · Proprietary	Anthropic implementation / Gemini 3.5 Flash grader	72.5%
GPT-6 AstraOpenAI · Proprietary	Anthropic implementation / Gemini 3.5 Flash grader	71%
Claude Opus 5.5Anthropic · Proprietary	Anthropic implementation / Gemini 3.5 Flash grader	64.4%
Claude Fable 5.1Anthropic · Proprietary	Anthropic implementation / Gemini 3.5 Flash grader	44.8%

View all 6 published reports

Claude Fable 5.291.6%

Anthropic implementation / Gemini 3.5 Flash grader

effort: max · 100 tasks; expert answer ranges; Gemini 3.5 Flash grader. Tool-enabled variant has container, standard libraries and image cropping. · Tools enabled

Claude Fable 5.273.4%

Anthropic implementation / Gemini 3.5 Flash grader

effort: max · 100 tasks; expert answer ranges; Gemini 3.5 Flash grader. Tool-enabled variant has container, standard libraries and image cropping. · No tools

Gemini 4 Carbon72.5%

Anthropic implementation / Gemini 3.5 Flash grader

effort: highest available · November 9 launch-table comparison; highest available effort unless explicitly noted. Preserve the benchmark track and model configuration; competitor figures are as reported by Anthropic. · No tools

Anthropic implementation / Gemini 3.5 Flash grader

effort: highest available · November 9 launch-table comparison; highest available effort unless explicitly noted. Preserve the benchmark track and model configuration; competitor figures are as reported by Anthropic. · No tools

Claude Opus 5.564.4%

Anthropic implementation / Gemini 3.5 Flash grader

effort: highest available · November 9 launch-table comparison; highest available effort unless explicitly noted. Preserve the benchmark track and model configuration; competitor figures are as reported by Anthropic. · No tools

Claude Fable 5.144.8%

Anthropic implementation / Gemini 3.5 Flash grader

effort: highest available · November 9 launch-table comparison; highest available effort unless explicitly noted. Preserve the benchmark track and model configuration; competitor figures are as reported by Anthropic. · No tools

Sources and original reports ↗