Hypersim
Progress Over Time
Interactive timeline showing model performance evolution on Hypersim
State-of-the-art frontier
Open
Proprietary
Hypersim Leaderboard
4 models
| Context | Cost | License | ||||
|---|---|---|---|---|---|---|
| 1 |  Alibaba Cloud / Qwen Team | 35B | — | — | ||
| 2 | Alibaba Cloud / Qwen Team | 27B | 262K | $0.30 / $2.40 | ||
| 3 | Alibaba Cloud / Qwen Team | 122B | — | — | ||
| 4 | Alibaba Cloud / Qwen Team | 236B | — | — |
Notice missing or incorrect data?
What is Hypersim?
Hypersim evaluates 3D grounding and depth understanding in synthetic indoor scenes.
Hypersim is a image benchmark evaluating models on spatial reasoning, 3d, and vision tasks. LLM Stats tracks 4 models on this benchmark, scored on a 0–100 scale. The current average is 0.1, with the leader at 0.1.
Compare leaders on the best AI for spatial reasoning, best AI for 3d and best AI for vision leaderboards.
Current leaders
Qwen3.5-35B-A3B from Alibaba Cloud / Qwen Team currently leads the Hypersim leaderboard with a score of 0.131 across 4 evaluated AI models.
FAQ
Common questions about the Hypersim benchmark and leaderboard.