TriViewBench - Level 1: leaderboard
Metric: Accuracy (%) over the 3,031 Level-1 questions (low object count, low 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 | 91.52 |
| 2 | GPT-4o | 86.41 |
| 3 | Qwen 3 VL 32B Instruct | 86.04 |
| 4 | Claude 3.7 Sonnet | 84.16 |
| 5 | Qwen 3 VL 8B Instruct | 79.54 |
| 6 | Qwen 3 VL 4B Instruct | 69.55 |
| 7 | Qwen 2.5 VL 32B Instruct | 67.21 |
| 8 | Qwen 2.5 VL 7B Instruct | 64.14 |
| 9 | InternVL3.5-8B | 40.84 |
Interactive version: theaggregate.ai/benchmark?slug=triviewbench-level-1 · How It Works · Data refreshed daily, snapshot 2026-09-29.