OmniFood-Bench - Nutritional Profile (Homemade Food): leaderboard

Metric: Nutritional-profile MAPE (%; mean absolute percentage error of the predicted macronutrient grams of Homemade Food images, lower is better and unbounded above; 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 April 2027. 6 models tracked.

Top models

#ModelScore
1GPT-5.149.17
2Claude Sonnet 461.22
3InternVL3.5-8B63.61
4Gemini 3 Flash114.28

Interactive version: theaggregate.ai/benchmark?slug=omnifood-bench-nutritional-profile-homemade-food · How It Works · Data refreshed daily, snapshot 2026-09-29.