Chatbot Arena (Text-to-Image): leaderboard
Human preference votes from blind pairwise comparisons of images generated from the same text prompt, measuring output quality, prompt adherence, and aesthetic appeal.
Metric: Elo. Source: arena.ai. 82 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | gpt-image-2.5-sunburst | 1425 | |
| 2 | gpt-image-2.5-flare | 1398 | |
| 3 | gpt-image-2 (medium) | 1383 | |
| 4 | grok-imagine-image-2.0 (canvas) | 1334 | |
| 5 | mai-image-2.6 | 1332 | |
| 6 | gemini-nano-banana-2.1 | 1328 | |
| 7 | reve-2.1 | 1302 | |
| 8 | grok-imagine-image-2.0 (20260801) | 1294 | |
| 9 | muse-image | 1274 | |
| 10 | reve-2.0 | 1269 |
Interactive version: theaggregate.ai/benchmark?slug=chatbot-arena-text-to-image · How It Works · Data refreshed daily, snapshot 2026-10-20.