Benchmarks · reasoning
BabyVision
Uses UniPat AI's first-party visual-reasoning leaderboard score and excludes the human baseline. Score: score. Collected by Jake Boggs' benchmark dashboard.
20 models · 20 results · data as of 2026-10-29 · unipat.ai ↗
Leaderboard
Qwen 3.8 Max max
All results
20 rows| # | Model | Effort | BabyVision |
|---|---|---|---|
| 1 | Qwen 3.8 Max | max | 91.3% |
| 2 | Fable 5 | max | 90.5% |
| 3 | GPT-5.6 Sol | max | 88.9% |
| 4 | Kimi K3 | max | 85.7% |
| 5 | GPT-5.5 | xhigh | 83.6% |
| 6 | Opus 4.8 | max | 81.2% |
| 7 | Muse Spark 1.1 | max | 76.3% |
| 8 | Qwen 3.7 Plus | high | 70.4% |
| 9 | Kimi K2.6 | max | 68.5% |
| 10 | Gemini 3.1 Pro | high | 51.6% |
| 11 | Gemini 3 Pro | high | 49.7% |
| 12 | GPT-5.4 | xhigh | 49.7% |
| 13 | Qwen 3.5 397B-A17B | max | 43.3% |
| 14 | Kimi K2.5 | max | 36.5% |
| 15 | GPT-5.2 | xhigh | 34.4% |
| 16 | Qwen 3 | high | 19.2% |
| 17 | GLM 4.6 V | low | 17.6% |
| 18 | Grok 4 | high | 16.2% |
| 19 | Opus 4.6 | max | 14.8% |
| 20 | Opus 4.5 | high | 14.2% |