PerceptionBench — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 16 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.6 Sol | 59.7 |
| 2 | Kimi K3 | 58.5 |
| 3 | Claude Fable 5 | 57.2 |
| 4 | Gemini 3.1 Pro (Preview) | 56.2 |
| 5 | GPT-5.5 | 55.8 |
| 6 | Seed 2.1 Pro | 55 |
| 7 | Gemini 3.5 Flash | 52 |
| 8 | Qwen 3.7 Plus | 51.1 |
| 9 | Qwen 3.5 397B A17B | 47.5 |
| 10 | Claude Opus 4.8 | 47.2 |
| 11 | Kimi K2.6 | 42.6 |
| 12 | Grok 4.5 | 41 |
| 13 | Gemma 4 31B | 40.7 |
| 14 | GLM-5V Turbo | 39.6 |
| 15 | MiniMax-M3 | 33.1 |
Interactive version: theaggregate.ai/benchmark?slug=perceptionbench · How It Works · Data refreshed daily, snapshot 2026-08-05.