Benchmark

Hypersim

Hypersim evaluates 3D grounding and depth understanding in synthetic indoor scenes.

Modality
image
Categories
spatial_reasoning, 3d, vision
Openness
unknown
Source
llm_stats
Reported scores
4

Reported scores

llm_stats

ModelOrganizationReported valueReportedEvidence
Qwen3 VL 235B A22B ThinkingQwen0.112025-09-22evidence
Qwen3.5-122B-A10BQwen0.1272026-02-24evidence
Qwen3.5-27BQwen0.132026-02-24evidence
Qwen3.5-35B-A3BQwen0.1312026-02-24evidence

Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.