EgoThinker-7B: benchmark results
Provider: Other. Released 2025-10-29. Access: Open.
Aggregate Rating 1551 ± 51, rank #1612 of 10556 rated models, from 20 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EgoActionBench - Temporal m-tIoU | 13.7 | Mean matched temporal IoU (0-100; all 1,407 episodes (Ego4D, | 77.8 |
| EgoActionBench - Temporal | 18.9 | Temporal Overall (0-100; mean of m-tIoU and micro-averaged F | 61.1 |
| EgoActionBench | 20.8 | EgoActionScore (0-100; mean of Temporal Overall and Semantic | 44.4 |
| EgoActionBench - Ego4D subset | 22.6 | EgoActionScore (0-100; mean of Temporal Overall and Semantic | 44.4 |
| EgoActionBench - EPIC-KITCHENS-100 subset | 20.4 | EgoActionScore (0-100; mean of Temporal Overall and Semantic | 33.3 |
| EgoActionBench - EgoLife subset | 18.9 | EgoActionScore (0-100; mean of Temporal Overall and Semantic | 33.3 |
| EgoActionBench - Semantic Hand | 33.7 | Hand score (0-100; hand-label agreement of the prediction se | 33.3 |
| EgoActionBench - Temporal F1 | 24.1 | Micro-averaged action F1 from one-to-one matching (0-100; al | 33.3 |
| EgoActionBench - Semantic Motion | 19.5 | Motion agreement judged by the language-model judge (0-100; | 30.6 |
| EgoActionBench - Semantic Verb | 19 | Verb agreement judged by the language-model judge (0-100; al | 27.8 |
| EgoActionBench - Semantic | 22.8 | Semantic Overall (0-100; applicable Hand, Verb, Objects, Too | 22.2 |
| EgoActionBench - Semantic Objects | 19.1 | Objects agreement judged by the language-model judge (0-100; | 16.7 |
Interactive version: theaggregate.ai/model?slug=egothinker-7b · How It Works · Data refreshed daily, snapshot 2026-11-17.