---
title: "图像编辑排行榜"
description: "图像编辑模型的排名。"
canonical: https://pirouter.ai/zh/rankings/imgedit
last-updated: 2026-10-15
---

# 图像编辑排行榜

> 图像编辑模型的排名。

- 更新于: 2026-09-30
- 52,294,902 票 · 55 个模型
- 分数 · 越高越好

## 完整排名

| Rank | Rank spread | Model | Vendor | Score | Votes |
|---|---|---|---|---|---|
| 1 | 1–1 | GPT-image 2.5 Sunburst | OpenAI | 1522 ±6 | 47289 |
| 2 | 2–2 | GPT-image 2.5 Flare | OpenAI | 1479 ±5 | 44792 |
| 3 | 3–3 | GPT-image 2 (medium) | OpenAI | 1461 ±3 | 293819 |
| 4 | 4–5 | Grok Imagine Image 2.0 (low) | xAI | 1427 ±5 | 36022 |
| 5 | 4–5 | MAI Image 2.6 | Microsoft AI | 1427 ±5 | 41512 |
| 6 | 6–7 | Muse Image | Meta | 1404 ±4 | 130653 |
| 7 | 6–7 | MAI Image 2.5 | Microsoft AI | 1401 ±4 | 190356 |
| 8 | 8–12 | Seedream 5.0 Pro | ByteDance | 1394 ±3 | 270020 |
| 9 | 8–12 | Grok Imagine Image Quality | xAI | 1391 ±6 | 38256 |
| 10 | 8–12 | [Nano Banana Pro](https://pirouter.ai/zh/models/google/gemini-3-pro-image) | Google | 1390 ±3 | 638610 |
| 11 | 8–12 | Chatgpt Image Latest High Fidelity | OpenAI | 1389 ±3 | 617941 |
| 12 | 8–12 | [Nano Banana 2](https://pirouter.ai/zh/models/google/gemini-3.1-flash-image) | Google | 1387 ±3 | 216640 |
| 13 | 13–15 | Reve 2.1 | Reve | 1374 ±6 | 20823 |
| 14 | 13–15 | GPT-image 1.5 High Fidelity | OpenAI | 1370 ±3 | 642272 |
| 15 | 13–16 | Qwen Image 2.1 | Alibaba | 1366 ±6 | 14617 |
| 16 | 15–17 | Reve 2.0 | Reve | 1359 ±6 | 19291 |
| 17 | 16–17 | Ideogram 4.5 | Ideogram | 1351 ±8 | 5441 |
| 18 | 18–19 | Uni 1.1 (max) | Luma | 1334 ±5 | 43451 |
| 19 | 18–19 | Grok Imagine Image | xAI | 1330 ±2 | 737207 |
| 20 | 20–20 | [Nano Banana 2 Lite](https://pirouter.ai/zh/models/google/gemini-3.1-flash-lite-image) | Google | 1314 ±5 | 65764 |
| 21 | 21–25 | Qwen Image 2.0 Pro 2026 06 22 | Alibaba | 1303 ±5 | 56781 |
| 22 | 21–25 | Hunyuan Image 3.0 Instruct | Tencent | 1303 ±3 | 331592 |
| 23 | 21–25 | Wan2.7 Image Pro | Alibaba | 1303 ±4 | 43992 |
| 24 | 21–25 | Seedream 4.5 | ByteDance | 1302 ±2 | 1266531 |
| 25 | 21–25 | Wan2.7 Image | Alibaba | 1301 ±4 | 44814 |
| 26 | 26–27 | Seedream 5.0 Lite | ByteDance | 1294 ±3 | 446300 |
| 27 | 26–27 | Gemini 2.5 Flash Image Preview | Google | 1293 ±2 | 11355727 |
| 28 | 28–30 | Seedream 4.2k | ByteDance | 1270 ±7 | 210076 |
| 29 | 28–30 | Reve V1.1 | Reve | 1262 ±3 | 672675 |
| 30 | 28–30 | Flux 2 (max) | Black Forest Labs | 1262 ±3 | 375959 |
| 31 | 31–32 | Kling Image O1 | Kling | 1251 ±4 | 143692 |
| 32 | 31–33 | Flux 2 Pro | Black Forest Labs | 1245 ±3 | 624352 |
| 33 | 32–34 | Qwen Image Edit | Alibaba | 1242 ±3 | 2022077 |
| 34 | 33–36 | Reve V1 | Reve | 1235 ±5 | 395314 |
| 35 | 34–36 | Qwen Image Edit 2511 | Alibaba | 1234 ±3 | 485308 |
| 36 | 34–37 | Wan2.6 Image | Alibaba | 1231 ±3 | 664189 |
| 37 | 37–39 | Flux 2 Flex | Black Forest Labs | 1225 ±3 | 443953 |
| 38 | 37–39 | Flux 2 Klein 9b | Black Forest Labs | 1225 ±3 | 540833 |
| 39 | 36–39 | Flux 2 Dev | Black Forest Labs | 1225 ±4 | 254450 |
| 40 | 40–41 | Seedream 4 High Res Fal | ByteDance | 1217 ±3 | 1229943 |
| 41 | 41–43 | P Image Edit | — | 1211 ±3 | 359965 |
| 42 | 40–43 | Seedream 4 Fal | ByteDance | 1210 ±6 | 152114 |
| 43 | 41–43 | Reve V1.1 Fast | Reve | 1207 ±3 | 553361 |
| 44 | 44–44 | Reve Edit Fast | Reve | 1198 ±4 | 230289 |
| 45 | 45–45 | Flux 2 Klein 4b | Black Forest Labs | 1188 ±3 | 541422 |
| 46 | 46–48 | Flux 1 Kontext (max) | Black Forest Labs | 1181 ±3 | 390877 |
| 47 | 46–48 | Wan2.5 I2i Preview | Alibaba | 1181 ±3 | 715740 |
| 48 | 46–48 | Flux 1 Kontext Pro | Black Forest Labs | 1177 ±3 | 6404790 |
| 49 | 49–49 | Flux 1 Kontext Dev | Black Forest Labs | 1149 ±3 | 3639915 |
| 50 | 50–51 | Seededit 3.0 | ByteDance | 1139 ±3 | 4918196 |
| 51 | 50–51 | GPT-image 1 | OpenAI | 1139 ±3 | 2870020 |
| 52 | 52–52 | GPT-image 1 Mini | OpenAI | 1124 ±3 | 682950 |
| 53 | 53–53 | Gemini 2.0 Flash Preview Image Generation | Google | 1081 ±2 | 4943024 |
| 54 | 54–54 | Bagel | ByteDance | 1027 ±6 | 13488 |
| 55 | 55–55 | Step1x Edit | StepFun | 998 ±4 | 155417 |

## FAQ

### 图像编辑最好的 AI 模型是哪个？

截至 2026-09-30，图像编辑榜第一为 OpenAI GPT-image 2.5 Sunburst，分数 1522，名次区间 1–1。

### 图像编辑最好的开放权重模型是哪个？

Qwen Image 2.1（Alibaba），全榜第 15 名。

### 这个榜单排的是什么？

图像编辑模型的排名。

### 多久更新一次？

有新模型发布或排名变化时更新；本榜最近一次更新为 2026-09-30。

数据：[LMArena leaderboard dataset](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset)，按 [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/) 许可使用；仅做筛选与推理档位合并，分数未改动。
查看原榜单: https://arena.ai/leaderboard/image-edit