GEdit-Bench
BAGEL-7B-MoT 6.52 
Step1X-Edit v1.2 7.58 leads since November 2025 Step1X-Edit v1.2 7.58 leads since November 2025 Image editing models follow real-world user edit instructions. A model judge scores semantic consistency, perceptual quality and an overall score.
Released with Step1X-Edit. The almanac files image generation under video-generation for now. BAGEL reports scores graded by GPT-4.1.
- Made by
- StepFun (Step1X-Edit team)
- Released
- April 2025
- Kind
- Image generation and editing
- Level
- Everyday
- Scored
- Graded by a model
- Answers
- Open-ended
- Tasks
- All public
- Last in a launch post
- May 2025, dropped before it saturated
- Last model added
- November 2025
- Scores refreshed
- November 12th, 2026
BAGEL-7B-MoT leads with 6.52
Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.
Step1X-Edit v1.2 leads with 7.58 across all independent runs
Best score per model across the official leaderboard, Epoch AI’s own runs and any board Epoch AI copied. Each bar and row names its source.
Step1X-Edit v1.2 leads with 7.58, claims included
Best score per model from every source, including labs’ own numbers and claims Epoch AI collected from reports and papers. Labs pick their own settings, so a claim can beat an independent run of the same model.
- 1 BAGEL-7B-MoT 6.52
Source: Labs’ model cards
- 1 Step1X-Edit v1.2 thinking + reflection 7.58
- 2
Qwen-Image-Edit-2509 7.56 - 3 Step1X-Edit v1.1 6.97
Source: Official GEdit-Bench leaderboard
- 1 Step1X-Edit v1.2 thinking + reflection 7.58
- 2 Qwen-Image-Edit-2509 7.56
- 3 Step1X-Edit v1.1 6.97
- 4 BAGEL-7B-MoT 6.52
Sources: Official GEdit-Bench leaderboard and labs’ model cards
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | BAGEL-7B-MoT | 6.52 | May 2025 | Lab model card |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Step1X-Edit v1.2 | 7.58 | thinking + reflection | Nov 2025 | Official leaderboard |
| 2 | Qwen-Image-Edit-2509 | 7.56 | Sep 2025 | Official leaderboard | |
| 3 | Step1X-Edit v1.1 | 6.97 | Jul 2025 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Step1X-Edit v1.2 | 7.58 | thinking + reflection | Nov 2025 | Official leaderboard |
| 2 | Qwen-Image-Edit-2509 | 7.56 | Sep 2025 | Official leaderboard | |
| 3 | Step1X-Edit v1.1 | 6.97 | Jul 2025 | Official leaderboard | |
| 4 | BAGEL-7B-MoT | 6.52 | May 2025 | Lab model card |
The benchmark's own leaderboard has 6 entries, 3 of them models we track. Switch the scores above to rank them together.
Number one over time 3 models have held number one across all independent runs since its release in April 2025 4 models have held number one with claims included since its release in April 2025
Lab claims count from the day the card came out, independent runs from the model’s release.
Step1X-Edit v1.2 7.58 Source: Official GEdit-Bench leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Sources: Official GEdit-Bench leaderboard and labs’ model cards
No launch post has listed it since May 2025
Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.
| Date | Launch post | Score | As the post writes it |
|---|---|---|---|
| May 20, 2025 | BAGEL-7B-MoT | 6.52 | GEdit-Bench-EN (O) 6.52, SC 7.36, PQ 6.83 |
One system card reports it
Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| May 20, 2025 | BAGEL-7B-MoT technical report | 6.52 | GEdit-Bench-EN (Full set) G_O, graded by GPT-4.1 (Table 7). G_SC 7.36, G_PQ 6.83. GEdit-Bench-CN G_O 6.50, G_SC 7.34, G_PQ 6.85 | Capabilities |
More in Vision
- MMMU In use 11.5K college-level multimodal questions across 30 subjects. Multiple-choice and open accuracy. GPT-5.1 85.4%
- MMMU-Pro In use Harder MMMU: college-level multimodal questions with more options and vision-only inputs to remove text shortcuts. GPT-5.6 Sol 84.6%
- CharXiv In use Chart understanding on real arXiv figures. Reasoning questions require synthesizing information from complex charts. Claude Fable 5 93.5%
- MATH-Vision (MathVision) In use 3,040 competition math problems with visual context. Answer accuracy. Qwen 3.8 Max 97.7%
- MMStar In use Models answer 1,500 human-curated image questions that cannot be solved without the image, across 6 core capabilities. Scored by accuracy. Qwen 3.8 Max 85.9%
- ChartQA In use Models answer human-written and generated questions about chart images that need visual and arithmetic reasoning. Scored by relaxed accuracy with a 5 percent numeric tolerance.
Command A Vision 90.9%