DiningBench - Nutrition Estimation: leaderboard
DiningBench (2026) evaluates vision-language models on multi-view user photos of restaurant dishes, testing dish classification against same-menu distractors, nutrition estimation and food visual question answering. This board is the mean absolute percentage error of the estimated calories, protein, carbohydrates and fat, averaged over the four nutrients; lower is better.
Metric: Mean absolute percentage error (%) of the estimated calories, protein, carbohydrates and fat (mean of the four MAPEs) over the 1,650 DiningBench nutrition-estimation dishes (multi-view restaurant photos; ground truth from merchant labels or ingredient-based estimates cross-checked against USDA FoodData Central); lower is better. Source: arxiv.org. Saturation forecast: Around March 2027. 29 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 24.45 | #119 |
| 2 | Gemini 3 Flash (Preview) | 25.21 | #152 |
| 3 | GPT-5 | 32.17 | #178 |
| 4 | O4 Mini | 33.95 | #368 |
| 5 | Qwen 3 VL 30B A3B Instruct | 37.35 | #895 |
| 6 | Gemini 2.5 Pro | 38.21 | #276 |
| 7 | GPT-4.1 | 38.6 | #505 |
| 8 | Qwen 3 VL 8B Instruct | 39.24 | #1029 |
| 9 | Qwen 2.5 VL 32B Instruct | 39.84 | #1062 |
| 10 | Qwen 3 VL 30B A3B (Thinking) | 40.22 | #719 |
| 11 | Qwen 2.5 VL 72B Instruct | 40.56 | #841 |
| 12 | Gemini 2.5 Flash | 41.6 | #463 |
| 13 | GPT-4o | 42.43 | #751 |
| 14 | Qwen 3 VL 4B Instruct | 42.55 | #1332 |
| 15 | Claude Sonnet 4.5 | 42.62 | #261 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=diningbench-nutrition-estimation · How It Works · Data refreshed daily, snapshot 2026-11-21.