Qwen: Qwen-Image-Edit
Qwen-Image-Edit is the image editing version of Qwen-Image, released by Alibaba's Qwen team. Built upon the 20B Qwen-Image model, it has been further trained to extend its unique text rendering capab…
siliconFlow/Qwen/Qwen-Image-Edit
Qwen-Image-Edit is the image editing version of Qwen-Image, released by Alibaba's Qwen team. Built upon the 20B Qwen-Image model, it has been further trained to extend its unique text rendering capabilities to image editing tasks, enabling precise text editing within images. Furthermore, Qwen-Image-Edit utilizes an innovative architecture that feeds the input image into both Qwen2.5-VL (for visual semantic control) and a VAE Encoder (for visual appearance control), achieving capabilities in both semantic and appearance editing. This allows it to support not only low-level visual appearance edits like adding, removing, or modifying elements, but also high-level visual semantic editing such as IP creation and style transfer, which require maintaining semantic consistency. The model has achieved state-of-the-art (SOTA) performance on multiple public benchmarks, establishing it as a powerful foundation model for image editing
Specifications & capabilities
- Input
- text, image
- Output
- image
- Published model ID
- Qwen-Image-Edit
- Capabilities
- imageOutput, multimodal, visionInput
Evaluations
1–1 / 1
1,240.875
- source id
- lmarena
- leaderboard name
- LMArena Image Edit Overall
- dimension
- multimodal
- metric
- Arena rating
- rank
- 42
- sample size
- 1999501
- evaluated at
- 2026-11-12T00:00:00.000Z
- source model id
- Qwen-Image-Edit
- confidence lower
- 1238.2464646898752
- confidence upper
- 1243.5042506813163
Original result ↗Published prices
1–1 / 1
$0.04
- Inference provider
- siliconFlow
- Platform
- SiliconFlow
- Unit
- image
- Currency
- USD
- Billing label
- Image
Provider specifications & access
1–3 / 3
- source model id
- publishers/qwen/models/qwen-image@qwen-image-edit
- hosting
- vertex-self-deployed
- source model id
- qwen/qwen-image-edit
- hosting
- nvidia-catalog
- source model id
- Qwen/Qwen-Image-Edit
- hosting
- siliconflow-hosted
Safety assessments
1–1 / 1
26 / 100
- source id
- stanfordFMTI
- canonical model id
- siliconFlow/Qwen/Qwen-Image-Edit
- taxonomy
- transparency
- hazard category
- transparency
- hazard name
- Organization transparency
- policy area
- Alibaba
- assessment type
- organization-level disclosure
- metric
- disclosure_score
- result label
- 26 / 100
- sample size
- 100
- updated at
- 2026-11-09T18:08:05.227Z
Source specification observations
1–1 / 1
- field
- context_length
- value
- 0K
- scope
- source-model
- source id
- siliconFlow
- confidence
- 1
Explore more models