FAVOR-Bench — leaderboard
Fine-grained Audio-Visual Object Recognition benchmark testing multimodal models on detailed object understanding.
Metric: Overall Score. Source: favor-bench.github.io. Status: saturation imminent. 21 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 1.5 Pro | 49.2 |
| 2 | Claude 3.7 Sonnet | 45.2 |
| 3 | GPT-4o | 40.6 |
| 4 | InternVL2.5-78B | 38.4 |
| 5 | InternVL2.5-2B | 18.7 |
Interactive version: theaggregate.ai/benchmark?slug=favor-bench · How the rankings work · Data refreshed daily, snapshot 2026-07-22.