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.
BabyVision
Higher is better · best effort per model
full table below
| # | Model | Effort | Score |
|---|---|---|---|
| 1 | Qwen 3.8 Max | max | 91.3% |
| 2 | Fable 5 | max | 90.5% |
| 3 | GPT-5.6 Sol | max | 88.9% |
| 4 | GPT-5.5 | xhigh | 83.6% |
| 5 | Opus 4.8 | max | 81.2% |
| 6 | Muse Spark 1.1 | max | 76.3% |
| 7 | Qwen 3.7 Plus | high | 70.4% |
| 8 | Kimi K2.6 | max | 68.5% |
| 9 | Gemini 3.1 Pro | high | 51.6% |
| 10 | Gemini 3 Pro | high | 49.7% |
| 11 | GPT-5.4 | xhigh | 49.7% |
| 12 | Qwen 3.5 397B-A17B | max | 43.3% |
| 13 | Kimi K2.5 | max | 36.5% |
| 14 | GPT-5.2 | xhigh | 34.4% |
| 15 | Qwen 3 | high | 19.2% |
| 16 | GLM 4.6 V | low | 17.6% |
| 17 | Grok 4 | high | 16.2% |
| 18 | Opus 4.6 | max | 14.8% |
| 19 | Opus 4.5 | high | 14.2% |