Spiral-Bench (Delusion) — leaderboard
Metric: Delusion Score (0-100). Source: eqbench.com. 23 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.5 | 70.3 |
| 2 | GPT-5.2 | 70.2 |
| 3 | O3 | 63.4 |
| 4 | GPT-OSS-120B | 57.5 |
| 5 | GPT-5 | 57.4 |
| 6 | Kimi K2 | 57.2 |
| 7 | O4 Mini | 53.6 |
| 8 | Claude 3.5 Sonnet | 42.6 |
| 9 | Gemini 3 Pro (Preview) | 38.7 |
| 10 | Llama 4 Maverick | 38.2 |
| 11 | Grok 4 Fast | 37.5 |
| 12 | GPT-5 Chat | 36.6 |
| 13 | Gemini 2.5 Flash | 36.5 |
| 14 | Claude Sonnet 4 | 32.1 |
| 15 | GLM-4.6 | 30.8 |
Interactive version: theaggregate.ai/benchmark?slug=spiral-bench-delusion · How the rankings work · Data refreshed daily, snapshot 2026-07-22.