spatial reasoning benchmark
Hypersim
Hypersim evaluates 3D grounding and depth understanding in synthetic indoor scenes.
Updated Aug 7, 2026
Published models4
Registry coverage4
MetricScore
EvidenceB
Hypersim leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | AC | 0.131 points | 100.0% | 4 | C | |
| 2 | AC | 0.13 points | 66.7% | 4 | C | |
| 3 | AC | 0.127 points | 33.3% | 4 | C | |
| 4 | AC | 0.11 points | 0.0% | 4 | C |
Score distribution
Top published rows on the benchmark's original scale.
Hypersim highlights
The top published results on this benchmark's own scale.
What is Hypersim?
Definition and scoring fields from the benchmark registry.
Hypersim evaluates 3D grounding and depth understanding in synthetic indoor scenes.
Scores are shown in points. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- Hypersim
- Modality
- image
- Primary category
- spatial reasoning
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- hypersim|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about Hypersim.
Which model scores highest on Hypersim?
Qwen3.5-35B-A3B is currently ranked first with 0.131 points.
What does Hypersim measure?
Hypersim evaluates 3D grounding and depth understanding in synthetic indoor scenes.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
4 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.