AgentIdeaBench - Active - Originality: leaderboard
Metric: Originality Critic Score (1-10). Source: moyunxiang.com. 63 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Opus 5 | 7.3 |
| 2 | GPT-5.6 Sol | 6.93 |
| 3 | Claude Opus 4.8 | 6.81 |
| 4 | GPT-5.6 Terra | 6.81 |
| 5 | Claude Opus 4.7 | 6.71 |
| 6 | GPT-5.5 | 6.7 |
| 7 | Claude Sonnet 5 | 6.7 |
| 8 | GPT-5.2 | 6.58 |
| 9 | GPT-5.4 | 6.51 |
| 10 | Gemini 3 Flash (Preview) | 6.33 |
| 11 | GLM-5.1 | 6.31 |
| 12 | GPT-5 | 6.29 |
| 13 | GPT-5.6 Luna | 6.28 |
| 14 | Gemini 3.1 Pro (Preview) | 6.2 |
| 15 | GPT-5.4 Mini | 6.2 |
Interactive version: theaggregate.ai/benchmark?slug=agentideabench-active-originality · How It Works · Data refreshed daily, snapshot 2026-09-19.