What is the AI2D leaderboard?

The AI2D leaderboard ranks 32 AI models based on their performance on this benchmark. Currently, Claude 3.5 Sonnet by Anthropic leads with a score of 0.947. The average score across all models is 0.872.

What is the highest AI2D score?

The highest AI2D score is 0.947, achieved by Claude 3.5 Sonnet from Anthropic.

How many models are evaluated on AI2D?

32 models have been evaluated on the AI2D benchmark, with 0 verified results and 32 self-reported results.

Where can I find the AI2D paper?

The AI2D paper is available at https://arxiv.org/abs/1603.07396. The paper details the methodology, dataset construction, and evaluation criteria.

Where can I find the AI2D dataset?

The AI2D dataset is available at https://allenai.org/data/diagrams.

What categories does AI2D cover?

AI2D is categorized under vision, multimodal, and reasoning. The benchmark evaluates multimodal models.

What is the best open-source model on AI2D?

Pixtral Large by Mistral AI is the top-ranked open-source model on AI2D, with a score of 0.938 (rank #4).

Which model offers the best value on AI2D?

Among models scoring within 10% of the leader, Qwen3 VL 8B Instruct from Alibaba Cloud / Qwen Team is the cheapest, at $0.08 per million input tokens with a score of 0.857.

How recent are the AI2D leaderboard results?

The AI2D leaderboard was last updated in June 2026 and currently includes 32 evaluated models.

All benchmarks

AI2D

AI2D is a dataset of 4,903 illustrative diagrams from grade school natural sciences (such as food webs, human physiology, and life cycles) with over 15,000 multiple choice questions and answers. The benchmark evaluates diagram understanding and visual reasoning capabilities, requiring models to interpret diagrammatic elements, relationships, and structure to answer questions about scientific concepts represented in visual form.

Claude 3.5 Sonnet from Anthropic currently leads the AI2D leaderboard with a score of 0.947 across 32 evaluated AI models.

Paper Implementation