Doubao-1.5-thinking-vision-pro: benchmark results
Provider: ByteDance.
Aggregate Rating 1695 ± 30, rank #371 of 10916 rated models, from 20 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| ReliableMath - Precision | 64.2 | Score (%) | 97.4 |
| LLM2014 Vision 2025-05 | 58.63 | Median Score | 92.9 |
| ReliableMath - Prudence | 0.5 | Score (%) | 84.2 |
| Embodied3DBench - Spatial Relation (3D) | 67.8 | Spatial Relation (3D) task (object-to-object and object-to-c | 83.3 |
| AEGIS (Academic Image Forensics) | 42.39 | Normalized Forensic Index (%): harmonic mean of the macro-F1 | 72 |
| AEGIS (Academic Image Forensics) - Textual Artifact Recognition | 75.44 | Macro-F1 (%) over two classes, telling whether the text regi | 68 |
| Embodied3DBench - Manipulation Trajectory (3D) | 34.7 | Manipulation Trajectory (3D) task (end-effector key trajecto | 66.7 |
| Embodied3DBench - Manipulation Trajectory (2D) | 22.6 | Manipulation Trajectory (2D) task (end-effector key trajecto | 58.3 |
| Embodied3DBench - Affordance (3D) | 34.8 | Affordance (3D) task (functional contact point and interacti | 50 |
| Embodied3DBench - Grasp Point (3D) | 47.4 | Grasp Point (3D) task (grasp points in metric 3D), normalize | 50 |
| Embodied3DBench - Grounding (3D) | 23.5 | Grounding (3D) task (metric 3D pointing and 3D bounding boxe | 50 |
| AEGIS (Academic Image Forensics) - Forgery Scope Discrimination | 35.31 | Macro-F1 (%) over three classes, telling whether an image is | 48 |
Interactive version: theaggregate.ai/model?slug=doubao-1-5-thinking-vision-pro · How It Works · Data refreshed daily, snapshot 2026-11-08.