EgoGPT-7B: benchmark results
Provider: Other.
Aggregate Rating 1495 ± 51, rank #2386 of 10556 rated models, from 16 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EgoActionBench - Semantic Motion | 20.3 | Motion agreement judged by the language-model judge (0-100; | 41.7 |
| EgoEsportsQA - Reasoning | 35.43 | Accuracy (%) on the reasoning-level EgoEsportsQA questions o | 37.5 |
| EgoEsportsQA | 34.73 | Overall accuracy (%) on the 1,745 EgoEsportsQA multiple-choi | 25 |
| EgoActionBench - Semantic Verb | 18.9 | Verb agreement judged by the language-model judge (0-100; al | 22.2 |
| EgoEsportsQA - Perception | 34.18 | Accuracy (%) on the perception-level EgoEsportsQA questions | 18.8 |
| EgoActionBench | 15.4 | EgoActionScore (0-100; mean of Temporal Overall and Semantic | 16.7 |
| EgoActionBench - EPIC-KITCHENS-100 subset | 16 | EgoActionScore (0-100; mean of Temporal Overall and Semantic | 16.7 |
| EgoActionBench - EgoLife subset | 15.9 | EgoActionScore (0-100; mean of Temporal Overall and Semantic | 16.7 |
| EgoActionBench - Temporal F1 | 16 | Micro-averaged action F1 from one-to-one matching (0-100; al | 16.7 |
| EgoActionBench - Semantic Hand | 24.1 | Hand score (0-100; hand-label agreement of the prediction se | 13.9 |
| EgoActionBench - Ego4D subset | 14.5 | EgoActionScore (0-100; mean of Temporal Overall and Semantic | 11.1 |
| EgoActionBench - Semantic | 20.4 | Semantic Overall (0-100; applicable Hand, Verb, Objects, Too | 11.1 |
Interactive version: theaggregate.ai/model?slug=egogpt-7b · How It Works · Data refreshed daily, snapshot 2026-11-17.