Image model leaderboard
Compare models within each evaluation task using original scores and configurations.
Ranked by original Elo within this task; higher is better. Elo is not comparable across tasks.
| 1 | GPT-Image-2.5 Sunburst OpenAI | 1226.7 Elo 95% CI1216.7–1236.7 | GPT Image 2.5 Sunburst (max) | 13,277 | 2026-11-05 |
| 2 | GPT-Image-2.5 Flare OpenAI | 1208.0 Elo 95% CI1198.0–1218.0 | GPT Image 2.5 Flare (max) | 12,616 | 2026-11-05 |
| 3 | GPT Image 2 OpenAI | 1187.9 Elo 95% CI1176.9–1198.9 | GPT Image 2 (high) | 13,096 | 2026-11-05 |
| 4 | Nano Banana 2.1 Google | 1175.1 Elo 95% CI1163.1–1187.1 | Nano Banana 2.1 | 6,421 | 2026-11-05 |
| 5 | Muse Image Meta | 1127.9 Elo 95% CI1116.9–1138.9 | Muse Image | 8,736 | 2026-11-05 |
| 6 | Nano Banana 2 Lite Google | 1108.9 Elo 95% CI1098.9–1118.9 | Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) | 15,355 | 2026-11-05 |
| 7 | Qwen-Image-3.0 Pro Alibaba | 1095.2 Elo 95% CI1083.2–1107.2 | Qwen-Image-3.0-Pro | 6,604 | 2026-11-05 |
| 8 | Qwen-Image-3.0 Alibaba | 1072.9 Elo 95% CI1060.9–1084.9 | Qwen-Image-3.0 | 6,336 | 2026-11-05 |