CharXiv Reasoning (with tools)Browse 296

CharXiv Reasoning (with tools)

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

Version not specifiedExact reported variant
Vision9 ranked models12 reported values3 reportsHigher is better

Top rankings

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

9
RankModelBest scoreBest reported setting
1Claude Opus 4.8Anthropic89.9%max effortMeta report1 value · 1 reportJul 9, 2026 · source
2Gemini 3.6 FlashGoogle89.4%Reported configurationGoogle report2 values · 2 reportsAug 13, 2026 · source
3Muse Spark 1.0Meta88.9%Reported configurationMeta report1 value · 1 reportJul 9, 2026 · source
4Gemini 3.7 FlashGoogle88.7%Reported configurationGoogle report1 value · 1 reportAug 13, 2026 · source
5Muse Spark 1.1Meta88.4%xhigh effortMeta report1 value · 1 reportJul 9, 2026 · source
6Claude Sonnet 5Anthropic88.3%Reported configurationGoogle report2 values · 2 reportsAug 13, 2026 · source
7Gemini 3.5 FlashGoogle84.9%Reported configurationGoogle report1 value · 1 reportJul 21, 2026 · source
8GPT-5.5OpenAI84.8%xhigh effortMeta report1 value · 1 reportJul 9, 2026 · source
9Gemini 3.1 ProGoogle83.2%Reported configurationGoogle report2 values · 2 reportsJul 21, 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.

With tools. Gemini's tool-enabled setup uses search and code execution. 1,000 questions; gpt-oss-120B high-reasoning judge; with code execution.

Method / source