MovieLLM: benchmark results
Provider: Other.
Aggregate Rating 1391 ± 51, rank #6478 of 10916 rated models, from 23 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MLVU-Test - Action Count | 21.7 | Multiple-choice accuracy (%) | 60.3 |
| MLVU-Dev - Generation | 3.94 | Generation score, GPT-4 judged (0-10) | 53.4 |
| MLVU-Test - Action Order | 20 | Multiple-choice accuracy (%) | 50 |
| MLVU-Test - Tutorial QA | 23.3 | Multiple-choice accuracy (%) | 45.6 |
| OpenVLM Video - MMBench-Video - Temporal Reasoning | 1.05 | GPT-4 Rating (0-3) | 34.1 |
| MLVU-Test - Anomaly Recognition | 25.6 | Multiple-choice accuracy (%) | 30.9 |
| OpenVLM Video - MMBench-Video - Relation Reasoning | 1.22 | GPT-4 Rating (0-3) | 27.3 |
| MLVU-Test - Generation | 3.52 | Generation score, GPT-4 judged (0-10) | 21.9 |
| MLVU-Test | 19.1 | Multiple-choice accuracy (%) | 20.6 |
| MLVU-Test - Topic Reasoning | 27.5 | Multiple-choice accuracy (%) | 20.6 |
| OpenVLM Video - MMBench-Video - Attribute Reasoning | 1.12 | GPT-4 Rating (0-3) | 14.8 |
| MLVU-Dev | 26.1 | Multiple-choice accuracy (%) | 10.5 |
Interactive version: theaggregate.ai/model?slug=moviellm · How It Works · Data refreshed daily, snapshot 2026-11-08.