Flourishing AI (FAI-G): leaderboard
Models answer single-turn questions about character, relationships, happiness, meaning, health, finances and faith, and judge models score how well each answer supports the user's overall flourishing. The general variant, not the FAI-C variant scored against a Christian worldview.
Metric: FAI-G overall flourishing score, June 2026 run (0-100); higher is better. Source: res.cloudinary.com. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-OSS-120B | 84 | #700 |
| 2 | Claude Sonnet 4.5 | 79 | #250 |
| 3 | GPT-5.1 | 79 | #229 |
| 4 | Qwen 3 235B A22B | 79 | #623 |
| 5 | Claude Sonnet 4 | 78 | #400 |
Interactive version: theaggregate.ai/benchmark?slug=flourishing-ai-fai-g · How It Works · Data refreshed daily, snapshot 2026-11-13.