Atlas

Benchmarks

← All benchmarks

PerceptionBench - Localization

Multimodal · 2026-07-16

The PerceptionBench localization category contains 330 questions that test whether a model can locate and anchor the correct region or object in an image.

Top models (higher is better)

ModelScore
GPT-5.6 Sol76.7
Claude Fable 570.4
Kimi K370.3
GPT-5.565.8
Opus 4.858.8
Gemini 3.1 Pro Preview52.7
Qwen3.7-Plus52.7
Gemini 3.5 Flash50.6
Doubao Seed 2.1 Pro50.0
Qwen3.5 397B A17B46.7
Kimi K2.645.2
Gemma 4 31B IT44.9
GLM 5V Turbo43.9
Grok 4.539.7
MiniMax M333.3
GLM-4.6V30.6
Loading Atlas data…