VisReason (Vision-Centric) - Cue Insight: leaderboard

VisReason Cue Insight is a conceptual-reasoning category of VisReason (2026): the model must combine textual and visual clues in an image to derive the answer. The score is accuracy in percent, higher is better.

Metric: Accuracy (%) on the Cue Insight category (conceptual reasoning: combine textual and visual clues in an image to derive the answer) 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)Low68.5#119
2GPT-5 MiniHigh60.4#364
3GPT-5Low53.2#178
4O4 Mini46#368
5GPT-5.2Thinking43.2#185
6Qwen 3 VL 8B Instruct34.3#1029
7Qwen 3 VL 235B A22B (Thinking)30.6#459
8GPT-5 Nano29.7#906
9Qwen 3 VL 30B A3B (Thinking)28.4#719
10GPT-4o27.1#751
11Qwen 3 VL 32B (Thinking)27#573
12Qwen 3 VL 8B (Thinking)25.2#910
13Qwen 2.5 VL 32B Instruct21.6#1062
14Gemini 2.5 Flash19.8#463
15Keye-VL-1.5-8BThinking16.6#1720

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-cue-insight · How It Works · Data refreshed daily, snapshot 2026-11-21.