PLCC - Overall: leaderboard
Metric: Mean category accuracy (%). Source: huggingface.co. 227 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 97 |
| 2 | Gemini 3.8 Flash | 97 |
| 3 | Gemini 3.7 Flash | 96.67 |
| 4 | Gemini 3.5 Flash | 96.33 |
| 5 | GPT-6 | 96.33 |
| 6 | GPT-5.5 (High) | 95.83 |
| 7 | Gemini 3.6 Flash | 95.5 |
| 8 | Claude Fable 5.1 | 95.17 |
| 9 | Claude Fable 5 | 95 |
| 10 | GPT-5.6 Sol (Medium) | 94.67 |
| 11 | GPT-5.5 (Low) | 94.5 |
| 12 | Claude Opus 5 | 92.17 |
| 13 | GPT-5.4 (High) | 92.17 |
| 14 | Gemini 2.5 Pro (Preview 06-05) | 92.17 |
| 15 | Muse Spark 1.3 | 91.83 |
Interactive version: theaggregate.ai/benchmark?slug=plcc-overall · How It Works · Data refreshed daily, snapshot 2026-09-19.