VisReason (Vision-Centric) - Multiple-Choice: leaderboard
Metric: Accuracy (%) on the 615 multiple-choice questions (option index extracted by regular expression) of VisReason, pooled over the 10 reasoning categories, 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 August 2028. 20 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro (Preview) (Low) | 47.8 |
| 2 | Qwen 3 VL 235B A22B (Thinking) | 40.8 |
| 3 | GPT-5.2 (Thinking) | 39.4 |
| 4 | GPT-5 Mini (Medium) | 36.4 |
| 5 | Qwen 3 VL 32B (Thinking) | 34.5 |
| 6 | GPT-5 Mini (High) | 34.3 |
| 7 | GPT-5 (Low) | 34 |
| 8 | GPT-5 Mini (Low) | 32.7 |
| 9 | Gemini 2.5 Flash | 31.5 |
| 10 | O4 Mini | 30.9 |
| 11 | Qwen 3 VL 30B A3B (Thinking) | 30.9 |
| 12 | Qwen 3 VL 8B (Thinking) | 30.6 |
| 13 | Qwen 3 VL 8B Instruct | 27.8 |
| 14 | GPT-4o | 25.2 |
| 15 | GPT-5 Nano | 23.6 |
Interactive version: theaggregate.ai/benchmark?slug=visreason-vision-centric-multiple-choice · How It Works · Data refreshed daily, snapshot 2026-10-07.