gdp.pdfBrowse 296

gdp.pdf

Can the model extract and reason over information in images or documents?

Version not specifiedExact reported variant
Vision13 ranked models21 reported values4 reportsHigher is better

Top rankings

One row per model, using its best reported score across effort settings.

13
RankModelBest scoreBest reported setting
1GPT-5.6 SolOpenAI40.0%Reported configurationGoogle DeepMind report2 values · 2 reportsSep 2, 2026 · source
2Claude Opus 5Anthropic37.0%Reported configurationGoogle DeepMind report1 value · 1 reportSep 2, 2026 · source
3Gemini 3.8 FlashGoogle35.0%Reported configurationGoogle DeepMind report1 value · 1 reportSep 2, 2026 · source
4Gemini 3.7 FlashGoogle34.0%Reported configurationGoogle DeepMind report2 values · 2 reportsSep 2, 2026 · source
5Claude Fable 5Anthropic29.8%Reported configurationOpenAI report1 value · 1 reportJul 9, 2026 · source
6GPT-5.6 TerraOpenAI29.0%Reported configurationGoogle DeepMind report3 values · 3 reportsSep 2, 2026 · source
7Claude Sonnet 5Anthropic28.0%Reported configurationGoogle DeepMind report2 values · 2 reportsSep 2, 2026 · source
8GPT-5.5OpenAI26.0%Reported configurationOpenAI report2 values · 2 reportsJul 9, 2026 · source
9GPT-5.6 LunaOpenAI22.7%Reported configurationOpenAI report1 value · 1 reportJul 9, 2026 · source
10Claude Opus 4.8Anthropic22.5%Reported configurationOpenAI report2 values · 2 reportsJul 9, 2026 · source
11Gemini 3.6 FlashGoogle22.0%Reported configurationGoogle report1 value · 1 reportAug 13, 2026 · source
12Gemini 3.1 Pro PreviewGoogle16.7%Reported configurationOpenAI report2 values · 2 reportsJul 9, 2026 · source

Effort curve

Every sourced cost-linked effort value for this exact version. Lines connect complete sweeps only.

0
No cost-linked effort sweep for this version.
Definition and comparison boundarypublic methodology

Can the model extract and reason over information in images or documents?

Visual perception and multimodal reasoning under the reported protocol. Higher is better. The value is the percentage reported in this lab's table.

Harness, tools, prompts, attempt count, and benchmark version can materially change the result. Compare within one reporting context.

No tools. Gemini and Claude Sonnet 5 cells are Google-computed; GPT-5.6 Terra and Muse Spark 1.2 come from the official public leaderboard. Expert PDF document comprehension; the model card labels this metric All pass rate.

Method / source