> ## Documentation Index
> Fetch the complete documentation index at: https://docs.qingbo.dev/llms.txt
> Use this file to discover all available pages before exploring further.
# GPT Image 2 Reverse
> Text-to-image and reference editing at a fixed price per image by output resolution
`gpt-image-2-rev` provides GPT Image 2 text-to-image generation and reference-image editing through an aggregator. Pricing is fixed by output resolution, making the per-image cost known before submission.
This is a reverse line, not the official OpenAI API. Use [`gpt-image-2`](/en/api-reference/image/gpt-image/gpt-image-2) for quality tiers, masks, transparent backgrounds, and output-format controls. QWave never switches between the two lines automatically.
## Quick start
```bash Generate an image theme={"system"}
curl -X POST https://www.qingbo.dev/v1/tasks \
-H "Authorization: Bearer $WAVE_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: image-rev-001" \
-d '{
"model": "gpt-image-2-rev",
"action": "generate",
"prompt": "Minimal perfume product photography, warm white background, soft studio light",
"aspect_ratio": "4:5",
"resolution": "1k",
"n": 1
}'
```
```bash Edit a reference image theme={"system"}
curl -X POST https://www.qingbo.dev/v1/tasks \
-H "Authorization: Bearer $WAVE_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: image-rev-edit-001" \
-d '{
"model": "gpt-image-2-rev",
"action": "edit",
"prompt": "Keep the subject and replace the background with a pale green studio",
"image_urls": ["https://cdn.example.com/source.png"],
"aspect_ratio": "1:1",
"resolution": "1k",
"n": 1
}'
```
A successful submission returns a `task_id`. Retrieve the result with [`GET /v1/tasks/{task_id}`](/en/api-reference/task/status), wait inline with [`Prefer: wait`](/en/api-reference/task/submit#in-request-waiting-prefer-wait), or configure a webhook.
## Request parameters
Must be `gpt-image-2-rev`.
* `generate`: create an image from text
* `edit`: edit reference images; requires `image_urls`
The image description or editing instruction.
Reference image URLs. Required for `edit`; up to 15 images. References do not add a fee.
One of `1:1`, `3:2`, `2:3`, `4:3`, `3:4`, `5:4`, `4:5`, `16:9`, `9:16`, `2:1`, `1:2`, `3:1`, `1:3`, `21:9`, or `9:21`.
Output resolution: `1k`, `2k`, or `4k`.
Number of images to generate. Value: `1`. Each request returns one image.
See [Submit a task](/en/api-reference/task/submit) for `callback_url`, `callback_events`, `Prefer: wait`, `Idempotency-Key`, and the maximum-cost header.
The reverse line does not support `quality`, `mask_url`, `background`, `output_format`, `output_compression`, `moderation` or `seed` — those belong to the official line. Any other parameter returns `400` and is not billed.
## Limits
| Condition | Limit |
| ---------------- | ---------------------------------------- |
| Per request | One image |
| `action: "edit"` | At least one reference image is required |
| Reference images | Up to 15 |
## Pricing
Successful images are billed at a **fixed price per image**, with a different rate at `1k`, `2k` and `4k`.
Rates are in `price_config` on `GET /v1/models` and in the console's Model Market. What a single call actually cost is the `cost` field on the task response — an integer quota at **500,000 quota = 1 USD**, not dollars.
`generate` and `edit` cost the same, reference images add no fee, and neither aspect ratio nor prompt length changes the price — **resolution is the only billing dimension, so the cost of a call is known before it is sent**. A `resolution` other than `1k`, `2k` or `4k` settles at the `1k` price.
Only a successfully generated image is billed. Failed and cancelled tasks, and tasks that return no usable image, are refunded in full.
## Response
```json Completed task theme={"system"}
{
"task_id": "taski_example",
"model": "gpt-image-2-rev",
"action": "edit",
"status": "completed",
"progress": "100%",
"result": {
"images": [
{
"url": ["https://cdn.example.com/result.png"]
}
]
},
"billing_status": "settled",
"cost": 4781
}
```
`result.images` carries the generated image. The full field reference is in [Task status](/en/api-reference/task/status).
## Official vs reverse
| | `gpt-image-2-rev` | [`gpt-image-2`](/en/api-reference/image/gpt-image/gpt-image-2) |
| ------------------------------ | ----------------------- | -------------------------------------------------------------- |
| Access | Aggregator reverse line | Official API line |
| Billing | Fixed by resolution | Actual token usage |
| Generation / reference editing | Supported | Supported |
| Quality tiers | Not supported | `low` / `medium` / `high` |
| Mask inpainting | Not supported | Supported |
| Background and output format | Not supported | Supported |
| Output count | 1 | 1 |
## Related
* [GPT Image Series](/en/api-reference/image/gpt-image/overview)
* [GPT Image 1 and 1.5 Official](/en/api-reference/image/gpt-image/gpt-image-1)
* [GPT Image 2 Official](/en/api-reference/image/gpt-image/gpt-image-2)
* [Submit Task](/en/api-reference/task/submit)
* [Query Task Status](/en/api-reference/task/status)
* [Task System](/en/docs/task-system)