Get task

Retrieve the status and result of a Qwen Image or Qwen Image 2.1 task using the task_id returned by Text to Image. This endpoint also retrieves existing Qwen Image editing tasks.

Supported model identifiers: Qubico/qwen-image-2.1 and Qubico/qwen-image.

Continue polling while the task is pending or processing. When status is completed, read output.image_url; if the task fails, inspect error. The output can be null before completion.

meta.usage.frozen reports reserved points, and meta.usage.consume reports consumed points. A successful text-to-image task costs 150000 points ($0.015); a successful legacy Qwen Image Edit task costs 250000 points ($0.025). Both have no remaining frozen points after successful completion. A newly accepted pending edit reserves 250000 points and consumes 0. An accepted pending-task cancellation releases the reserved points; it may be returned as failed with no consumed points.

Response input echoes submitted fields. Omitted numeric inputs can appear as 0 even though the generation defaults apply. For Qwen Image 2.1 these defaults are 1024 × 1024 pixels and 8 steps.

The examples below illustrate completed text-to-image tasks for both versions and a completed legacy image-edit task. Replace the example task ID and image URL with the values returned for your own task.

Qwen Image 2.1 uses the Turbo model for every request, with 1–16 steps (default and recommended: 8). Eight steps retain the official evaluated sampling schedule. Steps 1–7 keep the existing service-defined shorter schedule. Steps 9–16 use a service-defined schedule with deterministic piecewise linear interpolation over the nine official sigma anchors, including the final zero. These custom schedules are not an officially evaluated recipe and may reduce image quality; 9–16 steps are not guaranteed to outperform the recommended 8 steps. Omitted or non-positive step counts use 8; values above 16 are capped at 16. flow_shift and negative_prompt are accepted but ignored for 2.1. Qwen Image 2.1 supports text-to-image only; image editing uses Qubico/qwen-image.

At 1 step, some outputs can be heavily blurred and lose readable text. Use the default 8 steps for general use.

Legacy Qubico/qwen-image keeps 1–16 steps (default 8). Its text-to-image output area remains limited to 1,048,576 pixels, with each dimension 64–4096 pixels. For legacy text-to-image and image-edit, negative_prompt is accepted and encoded, but the current CFG=1 workflow does not provide classifier-free negative guidance.

Responses

🟢200Success

application/json

Bodyapplication/json

Examples

Qwen Image 2.1 completedQwen Image (existing model) completedQwen Image Edit completed

Illustrative response; task ID and output URL are placeholders.

{
  "code": 200,
  "data": {
    "task_id": "00000000-0000-4000-8000-000000000001",
    "model": "Qubico/qwen-image-2.1",
    "task_type": "txt2img",
    "status": "completed",
    "config": {
      "service_mode": "",
      "webhook_config": {
        "endpoint": "",
        "secret": ""
      }
    },
    "input": {
      "prompt": "A small ceramic teapot beside a window, clean product photograph, no text",
      "width": 1024,
      "height": 1024,
      "steps": 8,
      "seed": -1,
      "negative_prompt": ""
    },
    "output": {
      "image_base64": "",
      "image_url": "https://example.com/generated-image.png"
    },
    "meta": {
      "created_at": "2026-10-12T05:31:13.737522Z",
      "started_at": "2026-10-12T05:31:14.610515763Z",
      "ended_at": "2026-10-12T05:31:57.534876Z",
      "usage": {
        "type": "point",
        "frozen": 0,
        "consume": 150000
      },
      "is_using_private_pool": false
    },
    "detail": null,
    "logs": [],
    "error": {
      "code": 0,
      "message": "",
      "detail": null
    }
  },
  "message": "success"
}

Get task

Request

Request Code Samples

Responses