Video-CCAM-14B: benchmark results
Provider: Other.
Aggregate Rating 1549 ± 44, rank #1710 of 10916 rated models, from 18 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MLVU-Test - Action Count | 26.7 | Multiple-choice accuracy (%) | 82.4 |
| MLVU-Test - Ego Reasoning | 52.8 | Multiple-choice accuracy (%) | 79.4 |
| MLVU-Test - Plot QA | 56 | Multiple-choice accuracy (%) | 76.5 |
| MLVU-Dev - Generation | 4.01 | Generation score, GPT-4 judged (0-10) | 72.4 |
| MLVU-Dev | 63.1 | Multiple-choice accuracy (%) | 71.1 |
| MLVU-Test - Needle QA | 45 | Multiple-choice accuracy (%) | 67.6 |
| CG-Bench - Accuracy@IoU | 1.85 | Accuracy@IoU (%) | 61.9 |
| MLVU-Test | 42.9 | Multiple-choice accuracy (%) | 61.8 |
| MLVU-Test - Action Order | 24.3 | Multiple-choice accuracy (%) | 61.8 |
| MLVU-Test - Topic Reasoning | 79.1 | Multiple-choice accuracy (%) | 61.8 |
| MLVU-Test - Tutorial QA | 30.2 | Multiple-choice accuracy (%) | 57.4 |
| CG-Bench - Clue Grounding mIoU | 2.76 | Temporal Grounding mIoU (%) | 57.1 |
Interactive version: theaggregate.ai/model?slug=video-ccam-14b · How It Works · Data refreshed daily, snapshot 2026-11-08.