> ## Documentation Index > Fetch the complete documentation index at: https://docs.goenhance.ai/llms.txt > Use this file to discover all available pages before exploring further. # Qwen-Image 2.1 > Qwen-Image 2.1 in a single endpoint: text to image, editing with up to 10 reference images, transparent PNG output, and background removal. Omit `image_list` for text to image; pass it to edit. Always returns a single PNG. **Pricing** — flat per image by resolution tier. Text to image, editing and transparent output cost the same, and reference images add nothing: | image_size | Tokens | USD | |---|---|---| | 1K (about 1 MP, default) | 1 | $0.02 | | 2K (about 4 MP, native) | 4 | $0.08 | **Editing and reference images** - Refer to the images in the prompt as ``, `` … in `image_list` order. `image1` is the image being edited; the rest are references. - Up to 10 images at 1K. **At 2K the limit is 3**, because every reference is also encoded at 2K. - With `ratio: auto` (the default when editing) the output keeps the aspect ratio of `image1`. Choosing a ratio far from `image1` can shift the edit. - PNG references with transparency are read with their alpha channel. **Transparent background** — set `background` to `transparent` for a PNG with an alpha channel, for both text to image and editing. To remove the background of one image, send one image in `image_list` with `background: transparent` and no prompt. - Fully transparent pixels carry an arbitrary colour underneath. Composite the PNG over your own background before converting to JPEG, or the transparent area turns into a solid colour. - Transparent materials such as glass or lenses come out opaque. - For plain cut-outs, [Background Remover](/api-reference/image-api/background-remover) is cheaper and gives cleaner hair edges. **Timing** — about 30 to 60 seconds of processing for 1K, and about 3 to 7 minutes for 2K, plus queue time. ## OpenAPI ````yaml api-reference/image-api/openapi-qwen-image-2-1.json POST /api/v1/images/generations openapi: 3.1.0 info: title: GoEnhance API - Qwen-Image 3.0 description: Qwen-Image 3.0 version: 1.0.0 servers: - url: https://api.goenhance.ai security: [] paths: /api/v1/images/generations: post: tags: - ImageAPI summary: Qwen-Image 2.1 description: >- Qwen-Image 2.1 in a single endpoint: text to image, editing with up to 10 reference images, transparent PNG output, and background removal. Omit `image_list` for text to image; pass it to edit. Always returns a single PNG. **Pricing** — flat per image by resolution tier. Text to image, editing and transparent output cost the same, and reference images add nothing: | image_size | Tokens | USD | |---|---|---| | 1K (about 1 MP, default) | 1 | $0.02 | | 2K (about 4 MP, native) | 4 | $0.08 | **Editing and reference images** - Refer to the images in the prompt as ``, `` … in `image_list` order. `image1` is the image being edited; the rest are references. - Up to 10 images at 1K. **At 2K the limit is 3**, because every reference is also encoded at 2K. - With `ratio: auto` (the default when editing) the output keeps the aspect ratio of `image1`. Choosing a ratio far from `image1` can shift the edit. - PNG references with transparency are read with their alpha channel. **Transparent background** — set `background` to `transparent` for a PNG with an alpha channel, for both text to image and editing. To remove the background of one image, send one image in `image_list` with `background: transparent` and no prompt. - Fully transparent pixels carry an arbitrary colour underneath. Composite the PNG over your own background before converting to JPEG, or the transparent area turns into a solid colour. - Transparent materials such as glass or lenses come out opaque. - For plain cut-outs, [Background Remover](/api-reference/image-api/background-remover) is cheaper and gives cleaner hair edges. **Timing** — about 30 to 60 seconds of processing for 1K, and about 3 to 7 minutes for 2K, plus queue time. parameters: - name: Authorization in: header description: '' required: false example: '{{Authorization}}' schema: type: string requestBody: content: application/json: schema: type: object properties: model: type: string enum: - qwen-image-2-1 description: Model name. Must be `qwen-image-2-1`. prompt: type: string maxLength: 4000 description: >- What to generate, or the edit instruction. Refer to reference images as `` … ``. Required, except for background removal (one image in `image_list` with `background: transparent`). Any language is accepted, and text you ask for in the image is rendered as written. image_list: type: array items: type: object properties: url: type: string format: uri required: - url minItems: 1 maxItems: 10 description: >- Optional reference images for editing, [{"url": "https://..."}]. Up to 10 at 1K, up to 3 at 2K. The array order is the `` order; the first image is the one being edited. Reference images do not change the price. ratio: type: string enum: - auto - '1:1' - '2:3' - '3:2' - '3:4' - '4:3' - '9:16' - '16:9' - '21:9' default: auto description: >- Aspect ratio. `auto` follows the first reference image when editing and means 1:1 for text to image. image_size: type: string enum: - 1K - 2K default: 1K description: >- Resolution tier, also the price tier. 1K is about 1 megapixel, 2K about 4 megapixels (native). Case-insensitive. background: type: string enum: - opaque - transparent default: opaque description: >- `transparent` returns a PNG with an alpha channel. Does not affect pricing. seed: type: integer minimum: 0 maximum: 2147483647 description: Optional. Fixing the seed makes results more reproducible. custom_callback_url: type: string format: uri description: >- Optional. A publicly accessible HTTPS URL. When the task status changes (processing / success / failed), GoEnhance sends a POST request to this URL. The request body is identical to the response of GET /api/v1/jobs/detail. If your server does not respond with HTTP 200, the notification is retried up to 3 times, with a 3-second timeout per attempt. example: https://your-server.com/goenhance/callback required: - model example: model: qwen-image-2-1 prompt: >- A red panda barista making latte art in a cozy wooden cafe, warm morning light ratio: '16:9' image_size: 1K responses: '200': description: '' content: application/json: schema: type: object properties: code: type: integer msg: type: string data: type: object properties: img_uuid: type: string cost: type: number description: >- Tokens deducted for this request. This is the amount actually charged, so it already reflects any discount active on your account and can be lower than the listed price. Tokens are deducted when the task is accepted, and refunded automatically if the generation ends in failure. example: 1 required: - img_uuid - cost required: - code - msg - data examples: '1': summary: Success value: code: 0 msg: Success data: img_uuid: c12b656c-747a-44fd-9c80-add79b0c52d5 cost: 1 '2': summary: Insufficient tokens value: code: 100 msg: tokens is not enough quota: summary: API key quota exceeded value: code: 101 msg: 'API key quota exceeded: 4990 of 5000 tokens used (monthly)' headers: {} '401': description: '' content: application/json: schema: type: object properties: code: type: integer msg: type: string required: - code - msg headers: {} deprecated: false security: [] ````