# Image Editing on GEdit-Bench (GEdit-Bench Score)

Source: [Image Editing on GEdit-Bench (GEdit-Bench Score)](<https://www.sota2.com/research/sota/image-editing-on-gedit-bench-gedit-bench-score>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-05-12T08:24:17.767Z

Evaluation of image editing performance on the GEdit-Bench dataset, reporting the primary GEdit-Bench score.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Image Editing |
| Dataset (including any stated split) | GEdit-Bench |
| Metric | GEdit-Bench Score |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 19 of 19 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [GEdit-Bench Score](<https://www.sota2.com/research/sota/image-editing-on-gedit-bench-gedit-bench-score.md?metric=GEdit-Bench%20Score>): Higher is better; unit: SCALAR.

## Recorded results

| Method | GEdit-Bench Score | Result date | Report paper |
| --- | --- | --- | --- |
| w/ RPO-Gemini 3.1 Pro-ARR | 7.85 | Not recorded | 6a02b424dbc836b87d8246e4 |
| ARR-RPO | 7.85 | Not recorded | 6a02b424dbc836b87d8246e4 |
| ARR-RPO | 7.79 | Not recorded | 6a02b424dbc836b87d8246e4 |
| UniWorldv2 | 7.76 | Not recorded | 6a02b424dbc836b87d8246e4 |
| UniWorldv2 | 7.76 | Not recorded | 6a02b424dbc836b87d8246e4 |
| w/ RPO-GPT-5-ARR | 7.72 | Not recorded | 6a02b424dbc836b87d8246e4 |
| ARR-RPO | 7.72 | Not recorded | 6a02b424dbc836b87d8246e4 |
| w/ RPO-Qwen3vl-8B-ARR | 7.66 | Not recorded | 6a02b424dbc836b87d8246e4 |
| ARR-RPO | 7.66 | Not recorded | 6a02b424dbc836b87d8246e4 |
| ARR-RPO | 7.65 | Not recorded | 6a02b424dbc836b87d8246e4 |
| ARR-RPO | 7.63 | Not recorded | 6a02b424dbc836b87d8246e4 |
| Qwen-Image-Edit-2509 | 7.54 | Not recorded | 6a02b424dbc836b87d8246e4 |
| Qwen-Image-Edit-2509 | 7.54 | Not recorded | 6a02b424dbc836b87d8246e4 |
| Step1X-Edit | 6.97 | Not recorded | 6a02b424dbc836b87d8246e4 |
| Step1X-Edit | 6.97 | Not recorded | 6a02b424dbc836b87d8246e4 |
| Instruct-Pix2Pix | 3.68 | Not recorded | 6a02b424dbc836b87d8246e4 |
| Instruct-Pix2Pix | 3.68 | Not recorded | 6a02b424dbc836b87d8246e4 |
| AnyEdit | 3.21 | Not recorded | 6a02b424dbc836b87d8246e4 |
| AnyEdit | 3.21 | Not recorded | 6a02b424dbc836b87d8246e4 |

### w/ RPO-Gemini 3.1 Pro-ARR

GEdit-Bench Score: 7.85. Result ID: 6a02e39f47f34ce924e59f24.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Implementation](<https://github.com/OpenEnvision/AutoRubric-as-Reward>)

Recorded evaluation configuration: {"Base Generator":"Qwen-Image-Edit-2509"}. This record does not assert independent reproduction.

### ARR-RPO

GEdit-Bench Score: 7.85. Result ID: 6a02e39f47f34ce924e59f62.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Implementation](<https://github.com/OpenEnvision/AutoRubric-as-Reward>)

Recorded evaluation configuration: {"Variant":"w/ RPO-Gemini 3.1 Pro-ARR"}. This record does not assert independent reproduction.

### ARR-RPO

GEdit-Bench Score: 7.79. Result ID: 6a02e39f47f34ce924e59f61.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Implementation](<https://github.com/OpenEnvision/AutoRubric-as-Reward>)

Recorded evaluation configuration: {"Variant":"w/ RPO-Gemini 3.1 Pro"}. This record does not assert independent reproduction.

### UniWorldv2

GEdit-Bench Score: 7.76. Result ID: 6a02e39f47f34ce924e59f21.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c6a1e209131cea8df27.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### UniWorldv2

GEdit-Bench Score: 7.76. Result ID: 6a02e39f47f34ce924e59f5c.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7c6a1e209131cea8df27.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### w/ RPO-GPT-5-ARR

GEdit-Bench Score: 7.72. Result ID: 6a02e39f47f34ce924e59f23.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Implementation](<https://github.com/OpenEnvision/AutoRubric-as-Reward>)

Recorded evaluation configuration: {"Base Generator":"Qwen-Image-Edit-2509"}. This record does not assert independent reproduction.

### ARR-RPO

GEdit-Bench Score: 7.72. Result ID: 6a02e39f47f34ce924e59f60.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Implementation](<https://github.com/OpenEnvision/AutoRubric-as-Reward>)

Recorded evaluation configuration: {"Variant":"w/ RPO-GPT-5-ARR"}. This record does not assert independent reproduction.

### w/ RPO-Qwen3vl-8B-ARR

GEdit-Bench Score: 7.66. Result ID: 6a02e39f47f34ce924e59f22.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Implementation](<https://github.com/OpenEnvision/AutoRubric-as-Reward>)

Recorded evaluation configuration: {"Base Generator":"Qwen-Image-Edit-2509"}. This record does not assert independent reproduction.

### ARR-RPO

GEdit-Bench Score: 7.66. Result ID: 6a02e39f47f34ce924e59f5e.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Implementation](<https://github.com/OpenEnvision/AutoRubric-as-Reward>)

Recorded evaluation configuration: {"Variant":"w/ RPO-Qwen3vl-8B-ARR"}. This record does not assert independent reproduction.

### ARR-RPO

GEdit-Bench Score: 7.65. Result ID: 6a02e39f47f34ce924e59f5f.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Implementation](<https://github.com/OpenEnvision/AutoRubric-as-Reward>)

Recorded evaluation configuration: {"Variant":"w/ RPO-GPT-5"}. This record does not assert independent reproduction.

### ARR-RPO

GEdit-Bench Score: 7.63. Result ID: 6a02e39f47f34ce924e59f5d.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Implementation](<https://github.com/OpenEnvision/AutoRubric-as-Reward>)

Recorded evaluation configuration: {"Variant":"w/ RPO-Qwen3vl-8B"}. This record does not assert independent reproduction.

### Qwen-Image-Edit-2509

GEdit-Bench Score: 7.54. Result ID: 6a02e39f47f34ce924e59f20.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a6a1e209131cea83137.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Qwen-Image-Edit-2509

GEdit-Bench Score: 7.54. Result ID: 6a02e39f47f34ce924e59f5b.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a6a1e209131cea83137.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Step1X-Edit

GEdit-Bench Score: 6.97. Result ID: 6a02e39f47f34ce924e59f1f.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a551e209131cea82892.md>)

[Implementation](<https://github.com/stepfun-ai/Step1X-Edit>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Step1X-Edit

GEdit-Bench Score: 6.97. Result ID: 6a02e39f47f34ce924e59f5a.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a551e209131cea82892.md>)

[Implementation](<https://github.com/stepfun-ai/Step1X-Edit>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Instruct-Pix2Pix

GEdit-Bench Score: 3.68. Result ID: 6a02e39f47f34ce924e59f1d.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e641e209131cea3fa43.md>)

[Implementation](<https://www.sota2.com/timothybrooks.com/instruct-pix2pix>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Instruct-Pix2Pix

GEdit-Bench Score: 3.68. Result ID: 6a02e39f47f34ce924e59f58.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6e641e209131cea3fa43.md>)

[Implementation](<https://www.sota2.com/timothybrooks.com/instruct-pix2pix>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### AnyEdit

GEdit-Bench Score: 3.21. Result ID: 6a02e39f47f34ce924e59f1e.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77b91e209131cea74f9f.md>)

[Implementation](<https://dcd-anyedit.github.io/>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### AnyEdit

GEdit-Bench Score: 3.21. Result ID: 6a02e39f47f34ce924e59f59.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b424dbc836b87d8246e4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa77b91e209131cea74f9f.md>)

[Implementation](<https://dcd-anyedit.github.io/>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.