---
title: "Image editing: model rankings and source evidence · Undominated.ai"
canonical: https://undominated.ai/rankings/arena-image-edit/
description: "Image editing ranked by Arena score. Inspect exact model variants, source dates, measurement limits and downloadable evidence."
---

# Image editing: model rankings and source evidence · Undominated.ai

> Image editing ranked by Arena score. Inspect exact model variants, source dates, measurement limits and downloadable evidence.

[All rankings](/rankings/) / Images

# Image editing

Human preference ratings for the overall category of this Arena task. Scores from different boards are not interchangeable.

**63 entries** · Arena score · Higher is better

Source publication: 2026-11-04. Retrieved 2026-11-05.

Source: [Arena (lmarena-ai)](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset) · [CC-BY-4.0](https://creativecommons.org/licenses/by/4.0/).

Ranks use confidence bounds. Several models can share a rank; the rows within a shared rank are displayed by score.

Image editing · Arena score

| Significance rank | Model / evaluated variant | Arena score | 95% confidence interval | Votes |
| --- | --- | --- | --- | --- |
| 1 | GPT-Image-2.5 Sunburst openai | 1,517.3 | 1,512.58 – 1,522.03 | 114,468 |
| 2 | GPT-Image-2.5 Flare openai | 1,482.13 | 1,477.61 – 1,486.65 | 111,635 |
| 2 | GPT-Image-2.6 Sunburst (Medium) openai | 1,476.26 | 1,466.19 – 1,486.34 | 3,833 |
| 3 | MAI-Image-2.7-Preview microsoft-ai | 1,468.76 | 1,462.02 – 1,475.51 | 12,728 |
| 3 | GPT-Image-2.6 Flare (Medium) openai | 1,462 | 1,452.08 – 1,471.92 | 3,850 |
| 4 | GPT-Image-2 (Medium) openai | 1,460.09 | 1,457.09 – 1,463.09 | 359,999 |
| 7 | Grok Imagine Image 2.0 xai | 1,438.04 | 1,432.54 – 1,443.55 | 37,138 |
| 7 | Gemini Nano Banana 2.1 google | 1,427.31 | 1,421.51 – 1,433.12 | 12,853 |
| 8 | MAI-Image-2.6 microsoft-ai | 1,426.31 | 1,421.83 – 1,430.78 | 89,003 |
| 10 | Grok Imagine Image 2.0 (20260801) xai | 1,416.06 | 1,411.75 – 1,420.37 | 114,884 |
| 8 | Krea 3 Ultra krea | 1,411.14 | 1,400.26 – 1,422.02 | 2,918 |
| 11 | MAI-Image-2.5 microsoft-ai | 1,400.28 | 1,396.59 – 1,403.96 | 190,011 |
| 11 | Muse Image meta | 1,399.39 | 1,395.62 – 1,403.17 | 196,287 |
| 14 | Seedream 5.0 Pro bytedance | 1,390.94 | 1,387.74 – 1,394.14 | 342,188 |
| 13 | Grok Imagine Image (Quality, 20260519) xai | 1,390.73 | 1,385.12 – 1,396.34 | 38,203 |
| 14 | Gemini 3 Pro Image (nano-banana-pro, 2K) google | 1,390.2 | 1,387.73 – 1,392.68 | 706,552 |
| 14 | Gemini 3.1 Flash Image (nano-banana-2, Web Search) google | 1,386.55 | 1,383.55 – 1,389.55 | 283,121 |
| 14 | ChatGPT Image Latest (High Fidelity, 20251216) openai | 1,386.36 | 1,383.78 – 1,388.95 | 678,695 |
| 14 | Gemini 3 Pro Image Preview (nano-banana-pro) google | 1,385.22 | 1,382.37 – 1,388.06 | 540,524 |
| 20 | Reve 2.1 reve | 1,373.54 | 1,367.48 – 1,379.59 | 20,797 |
| 20 | GPT-Image-1.5 (High Fidelity) openai | 1,367.95 | 1,365.42 – 1,370.48 | 704,550 |
| 20 | Qwen-Image-2.1 alibaba | 1,363.71 | 1,357.81 – 1,369.61 | 41,482 |
| 22 | Reve 2.0 reve | 1,358.24 | 1,351.77 – 1,364.7 | 19,289 |
| 24 | Ideogram 4.5 ideogram | 1,346.64 | 1,342.22 – 1,351.06 | 87,673 |
| 25 | UNI-1.1 Max luma-ai | 1,333.85 | 1,328.98 – 1,338.72 | 43,467 |
| 25 | Grok Imagine Image xai | 1,328.13 | 1,325.74 – 1,330.53 | 844,208 |
| 27 | UNI-1.1 luma-ai | 1,315.98 | 1,312.77 – 1,319.2 | 238,890 |
| 27 | Gemini 3.1 Flash-Lite Image (nano-banana-2-lite) google | 1,312.98 | 1,308.27 – 1,317.7 | 65,603 |
| 29 | Seedream 4.5 bytedance | 1,302.64 | 1,300.53 – 1,304.74 | 1,406,120 |
| 29 | Qwen-Image-2.0-Pro (2026-06-22) alibaba | 1,302.46 | 1,297.82 – 1,307.1 | 56,688 |
| 29 | Wan 2.7 Image Pro alibaba | 1,302.18 | 1,298.1 – 1,306.26 | 43,970 |
| 29 | HunyuanImage 3.0 Instruct tencent | 1,301.9 | 1,298.59 – 1,305.21 | 331,417 |
| 29 | Wan 2.7 Image alibaba | 1,300.74 | 1,296.89 – 1,304.59 | 44,782 |
| 33 | Seedream 5.0 Lite bytedance | 1,294.7 | 1,292.11 – 1,297.28 | 520,751 |
| 34 | Gemini 2.5 Flash Image Preview (nano-banana) google | 1,291.51 | 1,289.52 – 1,293.5 | 11,489,917 |
| 36 | Seedream 4.0 (2K) bytedance | 1,266.75 | 1,259.85 – 1,273.65 | 210,026 |
| 36 | FLUX.2 [max] bfl | 1,261.49 | 1,258.36 – 1,264.62 | 375,773 |
| 36 | Reve v1.1 reve | 1,261.43 | 1,258.7 – 1,264.16 | 672,524 |
| 39 | Kling Image O1 kling | 1,250.45 | 1,246.79 – 1,254.11 | 143,679 |
| 39 | FLUX.2 [pro] bfl | 1,248.1 | 1,245.59 – 1,250.61 | 694,796 |
| 41 | Qwen-Image-Edit alibaba | 1,241 | 1,238.42 – 1,243.59 | 2,021,532 |
| 41 | Reve v1 reve | 1,234.52 | 1,229.55 – 1,239.49 | 395,056 |
| 42 | Qwen-Image-Edit (2511) alibaba | 1,233.2 | 1,230.45 – 1,235.95 | 484,921 |
| 42 | Wan 2.6 Image alibaba | 1,231.21 | 1,228.67 – 1,233.75 | 734,720 |
| 45 | FLUX.2 [flex] bfl | 1,224.55 | 1,221.43 – 1,227.66 | 443,862 |
| 45 | FLUX.2 [klein] 9B bfl | 1,224.37 | 1,221.51 – 1,227.23 | 540,673 |
| 45 | FLUX.2 [dev] bfl | 1,224.15 | 1,220.17 – 1,228.13 | 254,033 |
| 48 | Seedream 4.0 (High Res, fal) bytedance | 1,215.88 | 1,213.1 – 1,218.66 | 1,229,477 |
| 48 | P-Image-Edit Organization not supplied | 1,209.99 | 1,206.87 – 1,213.11 | 359,820 |
| 48 | Seedream 4.0 (fal) bytedance | 1,209.77 | 1,203.65 – 1,215.89 | 152,100 |
| 49 | Reve v1.1 Fast reve | 1,206.65 | 1,203.66 – 1,209.64 | 553,195 |
| 52 | Reve Edit Fast reve | 1,198.01 | 1,193.88 – 1,202.14 | 230,078 |
| 53 | FLUX.2 [klein] 4B bfl | 1,187.11 | 1,184.24 – 1,189.99 | 541,193 |
| 53 | FLUX.1 Kontext [max] bfl | 1,181.95 | 1,179.14 – 1,184.75 | 390,856 |
| 54 | Wan 2.5 Preview alibaba | 1,180.33 | 1,177.7 – 1,182.96 | 723,596 |
| 54 | FLUX.1 Kontext [pro] bfl | 1,177.05 | 1,174.38 – 1,179.72 | 6,404,206 |
| 57 | FLUX.1 Kontext [dev] bfl | 1,150.09 | 1,146.61 – 1,153.57 | 3,639,585 |
| 58 | SeedEdit 3.0 bytedance | 1,138.89 | 1,135.11 – 1,142.67 | 4,918,170 |
| 58 | GPT-Image-1 openai | 1,138.67 | 1,135.95 – 1,141.4 | 2,869,508 |
| 60 | GPT-Image-1 mini openai | 1,123.87 | 1,120.95 – 1,126.79 | 682,664 |
| 61 | Gemini 2.0 Flash Preview Image Generation google | 1,081.66 | 1,079.23 – 1,084.09 | 4,942,627 |
| 62 | BAGEL bytedance | 1,026.45 | 1,020.7 – 1,032.2 | 13,488 |
| 63 | Step1X-Edit stepfun | 996.02 | 991.47 – 1,000.58 | 155,418 |

## What this ranking can tell you

 - The population is the models in this source board, including variants and models without an available API price.
- Ranks use the published confidence bounds: one plus the number of models whose lower bound exceeds this model’s upper bound. Shared ranks are not proof of equal performance.

## Check the source

The original evidence is retained byte for byte. The downloadable board includes the source URLs, revision and SHA-256 hashes.

[Download this board and provenance →](/data/rankings/arena-image-edit.json)

 Source files and licences - [metadata · a7a159f557a1](/data/rankings/evidence/a7a159f557a1e594a5d41218fe8d53c63c63daa39b655e51d6fc5b482a17d059.json) · [upstream source ↗](https://huggingface.co/api/datasets/lmarena-ai/leaderboard-dataset)
- [licence · 88a46d472f35](/data/rankings/evidence/88a46d472f35274fb6b7a852c3d16c8ee8007d51b37f290161537bec26895ee4.txt) · [upstream source ↗](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/raw/8b6deb69ace01ee8dcda9e46eba7a5e8a808ec53/README.md)
- [rows · 3eb4615ba855](/data/rankings/evidence/3eb4615ba85527dfadf57a4e5a6628288b46741fd94df92991bce41a9977ff3d.parquet) · [upstream source ↗](https://huggingface.co/datasets/lmarena-ai/leaderboard-dataset/resolve/8b6deb69ace01ee8dcda9e46eba7a5e8a808ec53/image_edit/latest-00000-of-00001.parquet)

[Explore other model types →](/rankings/)[Read the benchmark guide →](/benchmarks/)

## Continue your investigation

 - [Understand the measurements](/benchmarks/)
- [Compare task performance](/best/)
- [Compare embedding models](/embeddings/)
- [Inspect ranking rules](/methodology/)