AI Image Editing Leaderboard
13 image-editing models, the same 18 photo edits, 5 runs each, judged with the model names hidden. Maintained by the OutfitGen Team and re-run as new models ship.
- Snapshot
- 2026-10 · v1.3
- Updated
- October 19, 2026
- Runs
- 1,170
- Source
- Lab data
Edits done right
Higher is better
Refusal rate
Lower is better
Median time per edit
Lower is better
Everyday fashion edits
| Model | ||||
|---|---|---|---|---|
| 1 | Flux 2 FlashBlack Forest Labs · Dec 2025 | 17 / 18 | ✓ 0%0 of 90 runs | 4.0 s |
| 2 | Flux 2 TurboBlack Forest Labs · Dec 2025 | 17 / 18 | ✓ 0%0 of 90 runs | 5.8 s |
| 3 | Seedream 5 FlashByteDance · Sep 2026 | 17 / 18 | ✓ 0%0 of 90 runs | 15.2 s |
| 4 | Seedream 5 ProByteDance · Jul 2026 | 17 / 18 | ✓ 0%0 of 90 runs | 27.2 s |
| 5 | Flux 2Black Forest Labs · Nov 2025 | 16 / 18 | ✓ 0%0 of 90 runs | 11.9 s |
| 6 | Seedream 5 LiteByteDance · Feb 2026 | 16 / 18 | ✓ 0%0 of 90 runs | 32.4 s |
| 7 | Qwen Image 3Alibaba · Jul 2026 | 15 / 18 | ✓ 0%0 of 89 runs | 14.5 s |
| 8 | Nano Banana 2Google · Feb 2026 | 15 / 18 | 21%19 of 90 runs | 11.3 s |
| 9 | Grok Imagine 2.0xAI · Aug 2026 | 14 / 18 | 17%15 of 89 runs | 13.2 s |
| 10 | GPT Image 2.5 Flare (medium)OpenAI · Sep 2026 | 14 / 18 | 22%20 of 90 runs | 15.1 s |
| 11 | GPT Image 2.5 Flare (high)OpenAI · Sep 2026 | 14 / 18 | 24%22 of 90 runs | 21.6 s |
| 12 | Nano Banana 2.1Google · Oct 2026 | 13 / 18 | 15%13 of 87 runs | 11.5 s |
| 13 | Muse ImageMeta · Sep 2026 | 12 / 18 | 26%23 of 89 runs | 16.5 s |
| Multi-model | OutfitGen EngineRoutes each edit across top models · tested October 2026 | 16 / 18 | ✓ 0%Delivered 90 of 90 runs | 16.7 s |
Ranked by edits done right, then refusal rate, then median time. OutfitGen Engine combines several models, so it is shown alongside the ranking. One edit is about 6 points on an 18-edit test, so treat a one-edit gap as close. Per-case results are in the source study.
Virtual try-on
13 models copy a garment from a product photo onto a person: 13 cases, 3 runs each, judged blind. Read the study.
| Rank | Model | Try-ons done rightof 13, higher is better | Garment copied rightof 13 | Same person keptof 13 |
|---|---|---|---|---|
| 1 | Nano Banana 2.1 | 13 / 13 | 13 | 13 |
| 2 | Seedream 5 Flash | 13 / 13 | 13 | 13 |
| 3 | GPT Image 2.5 Flare (high) | 13 / 13 | 13 | 13 |
| 4 | GPT Image 2.5 Flare (medium) | 12 / 13 | 12 | 13 |
| 5 | Muse Image | 12 / 13 | 12 | 13 |
| 6 | Nano Banana 2 | 11 / 13 | 12 | 13 |
| 7 | Seedream 5 Pro | 11 / 13 | 12 | 13 |
| 8 | Flux 2 | 6 / 13 | 8 | 10 |
| 9 | Qwen Image 3 | 6 / 13 | 6 | 12 |
| 10 | Flux 2 Flash | 5 / 13 | 6 | 11 |
| 11 | Flux 2 Turbo | 4 / 13 | 6 | 10 |
| 12 | FASHN Try-On v1.6 | 4 / 13 | 5 | 13 |
| 13 | Grok Imagine 2.0 | 0 / 13 | 0 | 13 |
| Multi-model | OutfitGen EngineRoutes each edit across top models · tested October 2026 | 13 / 13 | 13 | 13 |
Ranked by try-ons done right (garment, person and pose right in every run), then median time in the study session. FASHN Try-On v1.6 is a dedicated try-on model; the others are general image editors. Seedream 5 Lite was not in this study. OutfitGen Engine combines several models, so it is shown alongside the ranking.
Portrait and scene edits
4 models, 11 edits, 3 runs each. Read the study.
| Model | Face kepthigher is better | Unasked-for changesfewer is better | Outfit copied from a photocomplete runs |
|---|---|---|---|
| GPT Image 2.5 Flare (medium) | 9 of 9 runs | 0 of 11 | 2 of 3 runs |
| GPT Image 2.5 Flare (high) | 9 of 9 runs | 2 of 11 | 2 of 3 runs |
| Nano Banana 2 | 9 of 9 runs | 4 of 11 | 3 of 3 runs |
| Seedream 5 Pro | 9 of 9 runs | Scene kept, not relit | 3 of 3 runs |
Method
- Test set
- 18 everyday fashion edits on synthetic adult photos: outfit changes, new backdrops, swimwear, activewear and catalog lingerie.
- Runs
- Every model got the same photo and instruction 5 times, through each model's hosted API, with no tuned prompts.
- Edits done right
- Edits a reviewer rated correct in every delivered run, judging shuffled contact sheets before the model names were revealed. An edit refused every time counts as not done.
- Refusal rate
- Runs the model declined with a content-policy refusal, out of runs that returned an answer. Every request was ordinary, so every refusal is a false positive.
- Median time
- Median seconds from request to delivered image, measured in the same session for every model. Provider queues move absolute times; the order is the finding.
- Lab data only
- Every number comes from test photos made for these studies. No production traffic and no user photos are used.
How to cite
OutfitGen. (2026). AI Image Editing Leaderboard, snapshot 2026-10, version 1.3. OutfitGen Research. https://outfitgen.ai/research/leaderboard
@misc{outfitgen-leaderboard-2026-10,
title = {AI Image Editing Leaderboard},
author = {{OutfitGen}},
year = {2026},
note = {OutfitGen Research, snapshot 2026-10, version 1.3},
url = {https://outfitgen.ai/research/leaderboard}
}Free to reuse with attribution under CC BY 4.0.
Snapshots
2026-10 · v1.3 · October 19, 2026
Added the virtual try-on panel: 13 models, including the dedicated try-on model FASHN Try-On v1.6, copying 5 garments from product photos onto 9 photos of people, 13 cases, three runs each, judged blind. Nano Banana 2.1, Seedream 5 Flash and GPT Image 2.5 Flare (high) got all 13 right, as did OutfitGen Engine, shown alongside the ranking. The main table did not change.
2026-10 · v1.2 · October 19, 2026
Added OutfitGen Engine alongside the ranking: OutfitGen's multi-model system, which routes each edit across models, run on the same 18 edits five times each. It delivered all 90 requests, refused none and got 16 of 18 edits right under blind judging. Its median time is scaled to the September session through a same-day Nano Banana 2 control. No ranked row changed.
2026-10 · v1.1 · October 14, 2026
Added Nano Banana 2.1 (Google, released 6 October 2026) on the same 18 edits, five runs each, judged blind: 13 of 18 edits done right, 13 of 87 ordinary requests refused. Its median time is scaled to the September session through a same-day Nano Banana 2 control (41.8 s against the control's 41.0 s in October; 11.3 s for the control in September). No other row changed.
2026-09 · v1.0 · September 30, 2026
First snapshot: 12 models on the 18-edit everyday fashion set, plus the 4-model portrait and scene panel.
Want a model added? Write to the address on the about page.
OutfitGen is an AI photo editor for outfits, backgrounds and poses. These tests decide which models OutfitGen Engine trusts for which edits.
Try OutfitGen