FoodSense: leaderboard
Metric: Average over taste, smell, texture and sound of Lin's concordance correlation coefficient between the model's predicted 1-5 rating and the mean human rating (about 22 annotators per image) on FoodSense's 438 held-out test food images, zero-shot with the authors' sensory-assessment prompt, ratings parsed by regex; range -1 to 1, shown times 100; higher is better. Source: arxiv.org. Saturation forecast: Around January 2027. 5 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemma 3 27B (IT) | 13.6 |
| 2 | Qwen 2.5 VL 32B Instruct | 12.4 |
Interactive version: theaggregate.ai/benchmark?slug=foodsense · How It Works · Data refreshed daily, snapshot 2026-10-07.