qwen-image-2.1

Quick example

curl https://reapi.ai/api/v1/images/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen-image-2.1",
    "prompt": "A cute cartoon dragon sticker, bold outline",
    "background": "transparent",
    "resolution": "2K"
  }'

import requests

resp = requests.post(
    "https://reapi.ai/api/v1/images/generations",
    headers={
        "Authorization": "Bearer YOUR_API_KEY",
        "Content-Type": "application/json",
    },
    json={
        "model": "qwen-image-2.1",
        "prompt": "A cute cartoon dragon sticker, bold outline",
        "background": "transparent",
        "resolution": "2K",
    },
    timeout=30,
)
print(resp.json())

const r = await fetch("https://reapi.ai/api/v1/images/generations", {
  method: "POST",
  headers: {
    Authorization: "Bearer YOUR_API_KEY",
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    model: "qwen-image-2.1",
    prompt: "A cute cartoon dragon sticker, bold outline",
    background: "transparent",
    resolution: "2K",
  }),
});
console.log(await r.json());

package main

import (
    "bytes"
    "encoding/json"
    "fmt"
    "io"
    "net/http"
)

func main() {
    body, _ := json.Marshal(map[string]any{
        "model":      "qwen-image-2.1",
        "prompt":     "A cute cartoon dragon sticker, bold outline",
        "background": "transparent",
        "resolution": "2K",
    })
    req, _ := http.NewRequest("POST",
        "https://reapi.ai/api/v1/images/generations", bytes.NewReader(body))
    req.Header.Set("Authorization", "Bearer YOUR_API_KEY")
    req.Header.Set("Content-Type", "application/json")

    resp, _ := http.DefaultClient.Do(req)
    defer resp.Body.Close()
    out, _ := io.ReadAll(resp.Body)
    fmt.Println(string(out))
}

Authentication

Every call needs a Bearer token. Generate keys at reapi.ai/settings/apikeys.

Authorization: Bearer YOUR_API_KEY

Endpoint

POST /api/v1/images/generations
GET  /api/v1/tasks/{id}

Submission is async. The POST returns immediately with a task_id; poll the task endpoint until it completes. Polling does not consume credits. A 1K image typically takes seconds; 2K takes noticeably longer, up to a few minutes with reference images.

Request body

One image per request — there is no n.

Checked before anything is submitted or charged:

• aspect_ratio: "auto" needs image_urls
• mask_url needs exactly one image_urls entry
• mask_url cannot be combined with background: "transparent"
• output_format: "jpeg" cannot be combined with background: "transparent"
• at most 10 image_urls; every media field is a public HTTP(S) URL

Transparent output

Set background: "transparent" and Qwen Image 2.1 returns a PNG (or WebP) with a real alpha channel — no separate background-removal step. Describe the subject only: a prompt that mentions a scene, surface or backdrop may come back opaque. With a reference image, the same setting edits a transparent layer or cuts a subject out of an ordinary photo.

{
  "model": "qwen-image-2.1",
  "prompt": "Extract the sneaker as a clean product cutout",
  "image_urls": ["https://example.com/sneaker-on-street.jpg"],
  "background": "transparent",
  "output_format": "png"
}

Local edits with mask_url

Send exactly one reference image and a mask of the same aspect ratio (a mask of a different size is scaled to it). White marks the area to change, black is kept; a few pixels of feathering along the edge help the blend. In this mode the output follows the reference image's shape, so aspect_ratio and enhance_prompt are ignored.

{
  "model": "qwen-image-2.1",
  "prompt": "A green velvet armchair",
  "image_urls": ["https://example.com/living-room.jpg"],
  "mask_url": "https://example.com/living-room-mask.png"
}

content_filter — boolean, default true

Screens the request and output. Enabled by default; pass content_filter: false from an API key to turn it off. Requests from the playground always run with the filter on. A request the filter rejects fails and the reserved credits are refunded.

With content_filter: false, the output is stored in an isolated bucket and its URLs stay valid for 30 days — mirror the files if you need them longer.

Troubleshooting

Response envelope

Submit and poll share the same shape — only status and output fill in over time.

{
  "id": "task_019dfd44b7fd74168541552a3260a623",
  "model": "qwen-image-2.1",
  "status": "completed",
  "created_at": 1758600000,
  "output": {
    "image_urls": ["https://cdn.reapi.ai/media/tasks/.../0.png"]
  },
  "error": null
}

Poll GET /api/v1/tasks/{id} (see the Tasks reference) until status === "completed". output.image_urls holds the single generated image. Generated URLs expire — mirror them to your own storage for long-term use. Failures return { error: { code, message, request_id } }; the full catalog is at /docs/api/errors.

Pricing

Qwen Image 2.1 bills per delivered image, at a rate set by resolution:

credits = ceil(per_image_rate[resolution] × 1000)     1 credit = $0.001

1. Resolution is the only price dimension. 2K costs twice 1K; an omitted resolution bills as 1K.
2. Mode does not change the price. Text-to-image, image-to-image and masked edits cost the same at the same resolution.
3. Reference images are free. One or ten references, the price is the same.
4. Failed generations are refunded in full.

Live per-image rates are on the model page — that table is generated from the current price, so it is always the authoritative number.

Related

• Errors catalog
• Authentication
• Quickstart
• Qwen Image 3.0
• Qwen Image 2