Hypersim: leaderboard
Hypersim evaluates 3D grounding and depth understanding in synthetic indoor scenes.
Metric: Score (score_points). Source: arxiv.org. Source: benchmarklist.com. Saturation forecast: Around June 2027. 4 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3.5 35B A3B | 13.1 | #256 |
| 2 | Qwen 3.5 27B | 13 | #187 |
| 3 | Qwen 3.5 122B A10B | 12.7 | #173 |
| 4 | Qwen 3 VL 235B A22B | 11 | #238 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=hypersim · How It Works · Data refreshed daily, snapshot 2026-10-20.