P-Bench - OneIG: leaderboard
P-Bench scores text-to-image providers over the OneIG prompt set with P-Judge, the overall quality judge PrunaAI runs itself, alongside price per image and median generation time.
Metric: P-Judge Overall. Source: huggingface.co. 38 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Bria 4 Fibo | 56.74 | |
| 2 | Nano Banana 2 | 55.17 | |
| 3 | Nano Banana Pro | 55.11 | |
| 4 | P Image 2 Ideogram High 1K | 54.77 | |
| 5 | FLUX 2 Pro | 54.68 | |
| 6 | FLUX 2 Max | 54.57 | |
| 7 | P Image 2 Ideogram Medium 1K | 54.27 | |
| 8 | P Image 2 Ideogram High 2K | 54.09 | |
| 9 | Seedream 3 | 53.92 | |
| 10 | GPT Image 1.5 | 53.77 |
Interactive version: theaggregate.ai/benchmark?slug=p-bench-oneig · How It Works · Data refreshed daily, snapshot 2026-10-20.