BenchLeader

SpatialViz-Bench

Spatial visualisation: rotation, folding and animation.

As of 19 Sept 2026, Gemini 2.5 Pro leads SpatialViz-Bench on BenchLeader with 44.7%, ahead of o1 at 41.4%, across 8 model configurations with a published result.

Published by
Epoch AI Benchmarking Hub
Category
Multimodal
Index weight
Reference only
Models
8
Data as of
19 Sept 2026

CC BY 4.0 — Epoch AI, ‘AI Benchmarking Hub’, epoch.ai/benchmarks. Mirrored boards credit their original publishers.

What the test looks like

Mental rotation, paper folding and similar spatial tasks presented as images.

How it is scored

Overall score, as published.

What to keep in mind

Vision models only; small set.

8 of 8
#
1Gemini 2.5 ProGoogle44.7%54.52025-06-17
2o1OpenAI41.4%53.82024-12-17
3Gemini 2.5 FlashGoogle36.9%52.52025-06-17
4Llama 4 ScoutMetaopen34.2%40.22025-04-05
5Claude 3.7 SonnetAnthropic33.9%50.22025-02-24
6Qwen2.5-VL 72B InstructAlibabaopen ↗33.3%2024-09-19
7Qwen VlmaxAlibaba32.0%2024-01-18
8Llama 4 MaverickMetaopen31.8%43.42025-04-05

Cite as: BenchLeader, “SpatialViz-Bench leaderboard”, https://www.benchleader.com/benchmarks/spatialviz_bench, data as of 19 Sept 2026.

SpatialViz-Bench: questions

What does SpatialViz-Bench measure?
Mental rotation, paper folding and similar spatial tasks presented as images. Scores are reported in percent of tasks solved; higher is better.
Which AI model leads SpatialViz-Bench?
Gemini 2.5 Pro leads SpatialViz-Bench with 44.7% as of 19 Sept 2026, ahead of o1 at 41.4%.
How many models have SpatialViz-Bench results?
8 model configurations have a SpatialViz-Bench result on BenchLeader, all taken from Epoch AI Benchmarking Hub.
Who runs SpatialViz-Bench and how often is it updated?
SpatialViz-Bench is published by Epoch AI Benchmarking Hub. BenchLeader re-reads the published results every morning and records the date each result was published.
Does SpatialViz-Bench count toward the BenchLeader Index?
No. SpatialViz-Bench is shown for reference but left out of the composite index.