OmniFood-Bench - Advisory (Obesity): leaderboard
OmniFood-Bench asks vision-language models to act as a dietitian and decide whether a pictured meal is a normal, controlled or avoided intake for a patient managing obesity. The reference label follows the nutrient ranges the benchmark defines for that condition, and the board is scored as classification accuracy.
Metric: Advisory accuracy (%; share of images for which the model gives the annotated intake recommendation - normal, controlled or avoid - for a obesity profile, the label following the authoritative nutrient ranges for that condition; images of the MM-Food-100K subset; zero-shot, one standardized prompt per task; the open-weight models see all 1,208 samples and the proprietary models a 496-sample subset). Source: arxiv.org. Saturation forecast: Around June 2027. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4 | 34.27 | #400 |
| 2 | Gemini 3 Flash | 34.07 | #164 |
| 3 | GPT-5.1 | 30.04 | #229 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=omnifood-bench-advisory-obesity · How It Works · Data refreshed daily, snapshot 2026-11-13.