LLM Stats (BabyVision): leaderboard
LLM Stats mirror of BabyVision, a visual reasoning benchmark covering tasks designed to be solvable by young children but challenging for vision-language models.
Metric: Score (%). Source: llm-stats.com. 16 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | DeepSeek-V4.1-Flash-1017 | 92.5 | |
| 2 | MiniMax M3.1 Flash | 91 | |
| 3 | DeepSeek-V4.1-Flash | 89.6 | |
| 4 | Kimi K3 | 85.7 | #30 |
| 5 | Qwen3.8-27B | 85.6 | |
| 6 | Muse Spark Open | 80.2 | #54 |
| 7 | Muse Spark 1.1 | 76.3 | #52 |
| 8 | Seed 2.1 Pro | 73.7 | #67 |
| 9 | Qwen3.7-Plus | 70.4 | |
| 10 | Kimi K2.6 | 68.5 | #104 |
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-babyvision · How It Works · Data refreshed daily, snapshot 2026-10-20.