TriViewBench - Level 4: leaderboard
Metric: Accuracy (%) over the 4,231 Level-4 questions (high object count, high occlusion) of all three reasoning categories; three orthographic views (front, side, top) of synthetic 3D tower scenes, direct prompting with a concise answer within 128 tokens, greedy decoding; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 18 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.5 Flash | 65.94 |
| 2 | Claude 3.7 Sonnet | 54.9 |
| 3 | Qwen 3 VL 8B Instruct | 46.68 |
| 4 | Qwen 3 VL 32B Instruct | 43.02 |
| 5 | GPT-4o | 42.76 |
| 6 | Qwen 3 VL 4B Instruct | 40.75 |
| 7 | Qwen 2.5 VL 32B Instruct | 32.1 |
| 8 | InternVL3.5-8B | 28.43 |
| 9 | Qwen 2.5 VL 7B Instruct | 25.05 |
Interactive version: theaggregate.ai/benchmark?slug=triviewbench-level-4 · How It Works · Data refreshed daily, snapshot 2026-09-29.