VisReason (Vision-Centric) - Deductive Reasoning: leaderboard

VisReason Deductive Reasoning is a conceptual-reasoning category of VisReason (2026): the model must derive the logically entailed conclusion from the visual and textual information. The score is accuracy in percent, higher is better.

Metric: Accuracy (%) on the Deductive Reasoning category (conceptual reasoning: derive the logically entailed conclusion from the visual and textual information) of VisReason, answers scored by format (regular-expression match, GPT-5-mini judge for open-ended answers, IoU above 0.5 for boxes), zero-shot with chain-of-thought prompt templates, one per answer format; rows not shaded gray in the paper use explicit reasoning (GPT-5 and Gemini 3 Pro at low reasoning effort); higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 20 models tracked. Updated 7 Oct 2026.

Top models

#ModelSettingScoreOverall rank
1Gemini 3 Pro (Preview)Low53.7#119
2GPT-5.2Thinking48.9#185
3Qwen 3 VL 235B A22B (Thinking)45.2#459
4GPT-5 MiniMedium44.4#364
5Qwen 3 VL 32B (Thinking)39.2#573
6GPT-5Low38.8#178
7Qwen 3 VL 30B A3B (Thinking)37.3#719
8O4 Mini36.9#368
9Qwen 3 VL 8B Instruct34.1#1029
10Gemini 2.5 Flash34#463
11Qwen 3 VL 8B (Thinking)34#910
12GPT-4o32.1#751
13MiMo-VL-7B-RL25.4#1155
14GPT-5 Nano24.6#906
15Qwen 2.5 VL 32B Instruct23.5#1062

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=visreason-vision-centric-deductive-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-21.