Meta AI

BabyVision

A multimodal benchmark for fine-grained visual perception and grounded reasoning tasks.

In index
Multimodal
Best in multimodal
Source ↗

Models11

Top result82.0%Qwen3.8 Max

Top-3 spread5.7 pts

Year2026

Result and price

6 of 11 results

Results

11 results

#ModelResult

01Qwen3.8 Max82.0%

02Muse Spark Open80.2%

03Muse Spark 1.176.3%

04Seed 2.1 Pro73.7%

05Inkling 2 Small68.6%

06Qwen3.8-27B65.7%

07Seed 2.1 Turbo62.9%

08GLM-5.3-Flash53.4%

09dots3-note Preview50.0%

10Nemotron 3.5 Super VL 120B A12B29.6%

11Nemotron 3.5 Super VL 120B A12B NVFP429.5%