← Benchmarks
BabyVision
A benchmark for early-stage visual reasoning and perception on child-like vision tasks.
id babyvision · max 1 · 13 models reported
| # | Model | Score |
|---|---|---|
| 1 | DeepSeek-V4.1-Flash DeepSeek · open | 0.90 |
| 2 | Kimi K3 Moonshot AI · open | 0.86 |
| 3 | Qwen3.8-27B Alibaba Cloud / Qwen Team · open | 0.86 |