# Instruction-Based Image Editing on Reason50K Story Reasoning

Source: [Instruction-Based Image Editing on Reason50K Story Reasoning](<https://www.sota2.com/research/sota/instruction-based-image-editing-on-reason50k-story-reasoning>)

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

Data updated: 2026-05-14T06:44:58.041Z

Evaluation of model performance on instruction-based image editing tasks using metrics focused on visual alignment and multimodal reasoning (CLIP, MLLM, Ins-Align) on the Reason50K Story Reasoning dataset.

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 | Instruction-Based Image Editing |
| Dataset (including any stated split) | Reason50K Story Reasoning |
| Metric | CLIP Score |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 16 of 16 |
| 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

- [CLIP Score](<https://www.sota2.com/research/sota/instruction-based-image-editing-on-reason50k-story-reasoning.md?metric=CLIP%20Score>): Higher is better; unit: PERCENT.

- [MLLM Score](<https://www.sota2.com/research/sota/instruction-based-image-editing-on-reason50k-story-reasoning.md?metric=MLLM%20Score>): Higher is better; unit: PERCENT.

- [Instruction Alignment Score](<https://www.sota2.com/research/sota/instruction-based-image-editing-on-reason50k-story-reasoning.md?metric=Instruction%20Alignment%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | CLIP Score | Result date | Report paper |
| --- | --- | --- | --- |
| ReasonBrain | 30.1 | Not recorded | 68fa7ba71e209131cea89a6f |
| SmartFreeEdit | 22.9 | Not recorded | 68fa7ba71e209131cea89a6f |
| GPT Image 1.5 | 22 | Not recorded | 68fa7ba71e209131cea89a6f |
| MIGE | 21.6 | Not recorded | 68fa7ba71e209131cea89a6f |
| Qwen-Image-Edit | 21.5 | Not recorded | 68fa7ba71e209131cea89a6f |
| SmartEdit | 21.1 | Not recorded | 68fa7ba71e209131cea89a6f |
| MagicBrush | 20.9 | Not recorded | 68fa7ba71e209131cea89a6f |
| UltraEdit | 20.9 | Not recorded | 68fa7ba71e209131cea89a6f |
| Flux Kontext | 20.6 | Not recorded | 68fa7ba71e209131cea89a6f |
| MGIE | 20.1 | Not recorded | 68fa7ba71e209131cea89a6f |
| Bagel-Thinking | 19.8 | Not recorded | 68fa7ba71e209131cea89a6f |
| InstructPix2Pix | 19.6 | Not recorded | 68fa7ba71e209131cea89a6f |
| ChatGPT-4o | 19.5 | Not recorded | 68fa7ba71e209131cea89a6f |
| Qwen-Image-Edit | 18.9 | Not recorded | 68fa7ba71e209131cea89a6f |
| Bagel | 18.4 | Not recorded | 68fa7ba71e209131cea89a6f |
| PixWizard | 17.2 | Not recorded | 68fa7ba71e209131cea89a6f |

### ReasonBrain

CLIP Score: 30.1. Result ID: 6a056ea747f34ce924ebc373.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba71e209131cea89a6f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7ba71e209131cea89a6f.md>)

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

### SmartFreeEdit

CLIP Score: 22.9. Result ID: 6a056ea747f34ce924ebc36b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba71e209131cea89a6f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa78231e209131cea76f5d.md>)

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

### GPT Image 1.5

CLIP Score: 22. Result ID: 6a056ea747f34ce924ebc372.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba71e209131cea89a6f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a4b1e209131cea82449.md>)

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

### MIGE

CLIP Score: 21.6. Result ID: 6a056ea747f34ce924ebc36a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba71e209131cea89a6f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a311e209131cea81a79.md>)

Recorded evaluation configuration: {"Author":"Tian et al."}. This record does not assert independent reproduction.

### Qwen-Image-Edit

CLIP Score: 21.5. Result ID: 6a056ea747f34ce924ebc370.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba71e209131cea89a6f.md>)

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

Recorded evaluation configuration: {"mode":"finetuned"}. This record does not assert independent reproduction.

### SmartEdit

CLIP Score: 21.1. Result ID: 6a056ea747f34ce924ebc367.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba71e209131cea89a6f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa717b1e209131cea51882.md>)

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

### MagicBrush

CLIP Score: 20.9. Result ID: 6a056ea747f34ce924ebc365.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba71e209131cea89a6f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa732e1e209131cea5af49.md>)

[Implementation](<https://osu-nlp-group.github.io/MagicBrush>)

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

### UltraEdit

CLIP Score: 20.9. Result ID: 6a056ea747f34ce924ebc368.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba71e209131cea89a6f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76461e209131cea6c155.md>)

[Implementation](<https://www.sota2.com/ultra-editing.github.io>)

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

### Flux Kontext

CLIP Score: 20.6. Result ID: 6a056ea747f34ce924ebc36e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba71e209131cea89a6f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79a81e209131cea7eb79.md>)

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

### MGIE

CLIP Score: 20.1. Result ID: 6a056ea747f34ce924ebc366.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba71e209131cea89a6f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa721b1e209131cea548a3.md>)

Recorded evaluation configuration: {"Author":"Fu et al."}. This record does not assert independent reproduction.

### Bagel-Thinking

CLIP Score: 19.8. Result ID: 6a056ea747f34ce924ebc36d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba71e209131cea89a6f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a311e209131cea81aad.md>)

[Implementation](<https://bagel-ai.org/>)

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

### InstructPix2Pix

CLIP Score: 19.6. Result ID: 6a056ea747f34ce924ebc364.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba71e209131cea89a6f.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.

### ChatGPT-4o

CLIP Score: 19.5. Result ID: 6a056ea747f34ce924ebc371.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba71e209131cea89a6f.md>)

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

### Qwen-Image-Edit

CLIP Score: 18.9. Result ID: 6a056ea747f34ce924ebc36f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba71e209131cea89a6f.md>)

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

Recorded evaluation configuration: {"mode":"base"}. This record does not assert independent reproduction.

### Bagel

CLIP Score: 18.4. Result ID: 6a056ea747f34ce924ebc36c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba71e209131cea89a6f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a311e209131cea81aad.md>)

[Implementation](<https://bagel-ai.org/>)

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

### PixWizard

CLIP Score: 17.2. Result ID: 6a056ea747f34ce924ebc369.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7ba71e209131cea89a6f.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa772c1e209131cea7191b.md>)

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.
