Llama-3-MixSense-v1.1: benchmark results
Provider: Meta. Released 2024-09-03. Access: Open.
Aggregate Rating 1365 ± 36, rank #7294 of 10916 rated models, from 400 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenVLM MMT-Bench - MIA | 90 | Score (%) | 100 |
| OpenVLM MMT-Bench - Scene Recognition | 85 | Score (%) | 97.3 |
| OpenVLM MMT-Bench - Spot The Diff | 95 | Score (%) | 97.3 |
| OpenVLM MMT-Bench - Disaster Recognition | 100 | Score (%) | 96.8 |
| OpenVLM MME - Count | 180 | Score | 96.4 |
| OpenVLM MMT-Bench - 3D Indoor Recognition | 50 | Score (%) | 96.4 |
| OpenVLM MMT-Bench - Traffic Light Understanding | 100 | Score (%) | 96.4 |
| OpenVLM POPE | 89.7 | Overall (%) | 96.3 |
| OpenVLM MMT-Bench - Counting By Visual Prompting | 50 | Score (%) | 95.4 |
| OpenVLM MMT-Bench - Color Constancy | 71.4 | Score (%) | 94.2 |
| OpenVLM MMT-Bench - Traffic Participants Understanding | 65 | Score (%) | 92.5 |
| OpenVLM SEEDBench IMG - Instance Interaction | 78.4 | Accuracy (%) | 91.8 |
Interactive version: theaggregate.ai/model?slug=llama-3-mixsense-v1-1 · How It Works · Data refreshed daily, snapshot 2026-11-08.