InsightVQA - Emotion Perception: leaderboard
InsightVQA (2026) is a hierarchical visual question answering benchmark on emotion-annotated images drawn from six public sources. This board is accuracy (%) on single-choice emotion-category and valence recognition questions with three distractors; higher is better.
Metric: Accuracy (%) on single-choice emotion-category and valence recognition questions (the answer plus three LLM-written distractors) in InsightVQA-Bench, the held-out test split (a tenth of the data, about 30,000 questions) of emotion-annotated images from six public sources; higher is better. Source: arxiv.org. Saturation forecast: Not forecast. 14 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 2.5 VL 7B Instruct | 57.95 | #1833 |
| 2 | Gemini 2.5 Flash | 56.83 | #453 |
| 3 | Claude 3.7 Sonnet | 56.37 | #458 |
| 4 | Qwen 2.5 VL 72B Instruct | 53.83 | #846 |
| 5 | Qwen 2.5 VL 32B Instruct | 52.8 | #1053 |
| 6 | DeepSeek V3.2 | 51.2 | #406 |
| 7 | GPT-4o | 50.63 | #741 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=insightvqa-emotion-perception · How It Works · Data refreshed daily, snapshot 2026-11-13.