qwen-image/qwen-image-edit-2511/lora
LoRA-based visual editing model offering structure-aware asset transformation for creative pros
Qwen Image 3.0 Edit takes one to three source images plus a written instruction and returns the edited result. These controls shape each Qwen Image 3.0 Edit run.
| Parameter | Required | Type | Default | Range / Options | Description |
|---|---|---|---|---|---|
image_urls* | Yes (*) | Array | Sample image | 1-3 images | The source image(s) to edit; the model keeps the rest of the picture. |
prompt* | Yes (*) | String | Example instruction | Up to ~4,500 tokens | Plain-language edit instruction. Quote any text to render it verbatim. |
resolution | No | String | 2k | 1k, 2k | Output detail tier: 1k or 2k. |
aspect_ratio | No | String | 1:1 | 1:1, 16:9, 9:16, 4:3, 3:4 | Output aspect ratio for the edited image. |
enable_prompt_expansion | No | Boolean | true | true, false | Enrich a short instruction before editing; turn off when it is already precise. |
seed | No | Integer | -1 | -1 or 0-2147483647 | Fix a seed to reproduce an edit; use -1 for fresh variations. |
Qwen Image 3.0 Edit is billed at $0.028 per output image at 1K or 2K, plus $0.0028 per input image you provide. For a batch, cost is the per-image rate times the number of outputs plus the input-image charge, so Qwen Image 3.0 Edit only charges for what you generate.
LoRA-based visual editing model offering structure-aware asset transformation for creative pros
OpenAI's GPT Image 2 Image Edit: Image-to-image edits with precise text control and in-out painting
Advanced concept-driven image editing with unified segmentation and detection for creators.
Change an image’s aspect ratio cleanly with Ideogram 3 Reframe.
Advanced AI editing merges scenes and styles with precise structure control for designers.
FLUX 3 Image is a multimodal image model with reference guidance and readable text
Qwen Image 3.0 Edit reworks an image you upload based on a plain-language instruction. Instead of masking and layering by hand, you describe the change - swap a background, recolor an object, restage a product, or restyle a portrait - and Qwen Image 3.0 Edit applies it while keeping the rest of the picture consistent.
You can supply one to three source images. Qwen Image 3.0 Edit reads them together, which is useful for combining elements or keeping a subject consistent across a change. Add a clear instruction describing the result you want.
Be specific about what should change and what should stay. Name the object, the new look, and anything to preserve, for example "replace the sky with a sunset, keep the building unchanged." Put any on-image wording in quotes so Qwen Image 3.0 Edit renders it verbatim.
Qwen Image 3.0 Edit offers a 1K or 2K detail tier and aspect ratios of 1:1, 16:9, 9:16, 4:3, and 3:4. Choose 1K for quick iterations and 2K for delivery-ready detail; check the current RunComfy panel for the exact options.
Yes. Set a fixed seed to reproduce a Qwen Image 3.0 Edit result with the same image and instruction, or use -1 to get a fresh variation each run. The optional prompt expansion setting enriches short instructions before editing.
Qwen Image 3.0 Edit is billed at $0.028 per output image for both 1K and 2K, plus $0.0028 for each input image you provide. For a batch, the total is the per-output rate times the number of results plus the input-image charge.
Yes. Prototype an edit in the RunComfy UI, then call the same Qwen Image 3.0 Edit model via the RunComfy HTTP API with identical parameters. You do not need to host or scale the model yourself, and the same inputs produce comparable results.
RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Models, enabling artists to harness the latest AI tools to create incredible art.
