hunyuan-turbos-vision: benchmark results
Provider: Tencent.
Aggregate Rating 1555 ± 40, rank #1620 of 10916 rated models, from 11 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EPIC-Bench - Visual Matching | 36.14 | Navigation score (0-100) on visual matching (localizing an o | 50 |
| EPIC-Bench - Contact Relationship | 22.02 | Manipulation score (0-100) on contact relationship (localizi | 22.7 |
| EPIC-Bench - Embodied Compositional Attributes | 25.58 | Target localization score (0-100) on the embodied compositio | 19.3 |
| EPIC-Bench - Spatial-Related Attributes | 33.33 | Target localization score (0-100) on the spatial-attribute t | 19.3 |
| EPIC-Bench - Placement Region | 26.69 | Manipulation score (0-100) on placement region (where an obj | 15.9 |
| EPIC-Bench - Feasible Path | 17.88 | Navigation path score (0-100) on feasible path planning (a p | 14.8 |
| EPIC-Bench | 24.61 | Average score (0-100) over all samples, each sample scored b | 13.6 |
| EPIC-Bench - Affordance Region | 0.67 | Manipulation score (0-100) on affordance region localization | 10.2 |
| EPIC-Bench - Basic Attributes | 24.72 | Target localization score (0-100) on the basic-attribute tas | 6.8 |
| EPIC-Bench - Ground Detection | 4.49 | Navigation score (0-100) on ground detection (localizing tra | 6.8 |
| LLM2014 Vision 2025-05 | 20.51 | Median Score | 0 |
Interactive version: theaggregate.ai/model?slug=hunyuan-turbos-vision · How It Works · Data refreshed daily, snapshot 2026-11-08.