Eagle2.5-8B: benchmark results
Provider: Other.
Aggregate Rating 1530 ± 43, rank #1953 of 10916 rated models, from 16 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| TimeBlind | 20.2 | Instance accuracy I-Acc (%) over TimeBlind's 600 instances ( | 61.9 |
| Molmo2-VideoCount | 28.9 | Exact-count accuracy (%; 533 validation queries) | 55 |
| MMMU Benchmark | 55.8 | Validation Score | 53.7 |
| TimeBlind - Accuracy | 64.7 | Standard accuracy (%) over all 2,400 TimeBlind video-questio | 52.4 |
| AnyGroundBench - Animal TVG | 24.8 | tIoU@0.3 (%; share of queries whose temporal IoU is at least | 50 |
| Molmo2 Human Preference | 1019 | Elo rating (bootstrap median, Bradley-Terry; pairwise human | 50 |
| AnyGroundBench - Public Security TVG | 47.6 | tIoU@0.3 (%; share of queries whose temporal IoU is at least | 42.9 |
| AnyGroundBench - Surgery TVG | 15.2 | tIoU@0.3 (%; share of queries whose temporal IoU is at least | 42.9 |
| Molmo2 Human Preference - Captioning | 987 | Elo rating (bootstrap median, Bradley-Terry; pairwise human | 40 |
| Molmo2-CapTest | 22.8 | Caption F1 (%; LLM-judged statement precision and recall aga | 35 |
| AnyGroundBench - Industry SVG | 2.36 | sIoU@0.3 (%; share of queries whose mean spatial IoU is at l | 28.6 |
| AnyGroundBench - Industry TVG | 7.69 | tIoU@0.3 (%; share of queries whose temporal IoU is at least | 28.6 |
Interactive version: theaggregate.ai/model?slug=eagle2-5-8b · How It Works · Data refreshed daily, snapshot 2026-11-08.