Name: Text Layer Extraction API: Editable Text from Images
Brand: magicapi
SKU: api-magicapi-graphic-text-layer-extraction-ideogram-ai-layerize
Availability: InStock
Author: API Market

Shashank Agarwal

# 🔤 Text Layer Extraction API - Editable Text from Flat Graphics (Ideogram Layerize)

## Overview
Ideogram Layerize takes a flat graphic, such as a banner, card or social post, removes the text from it and returns that text as structured, editable layers. Its developer describes it as a way to get text layers back for editing and recomposing a design.

Send the graphic as `flat_graphic_image`. Each request returns a list of two URLs: first a JSON layer manifest that describes every text element (content, position, size, font and color), then a PNG of the graphic with the text removed.

## Features
- **Text removal**: returns the background of the graphic with the text taken out, ready for new text.
- **Structured text layers**: every text element comes back with its words, position (`x`, `y`, `width`, `height`), font size, line height, alignment, angle, color and font effects such as `bold` or `all_caps`.
- **Font matching**: each text item lists the font file the model matched and a short list of candidate fonts.
- **Text roles**: each item is tagged with a style class such as `h1`, `body` or `small`.
- **Positioned HTML**: the manifest includes `text_html`, an absolutely positioned HTML version of the extracted text that you can overlay on the background image.
- **Font overrides**: set `font_name_h1`, `font_name_h2`, `font_name_body` and `font_name_small` to choose the font for each text role.
- **Prompt and seed**: an optional `prompt` guides the layerization; a `seed` makes a result reproducible.

## Use Cases
1. **Editing text in finished graphics**: change a headline, price or date on a banner when the original design file is missing.
2. **Localization**: extract the text, translate it and render it back over the clean background.
3. **Template creation**: turn a flat design into a reusable template with an empty background and known text positions.
4. **Design-to-code**: use `text_html` and the item coordinates as a starting point for an HTML version of a graphic.

## Visual Examples
### Example Input
`flat_graphic_image`: a product card with a "NEW" label, the heading "api.market", a tagline and three small labels (Images, Video, Speech).

![Input graphic: a dark product card with a NEW label, the api.market heading, a tagline and Images, Video and Speech labels](https://blog.api.market/wp-content/uploads/2026/10/html-to-image-output-api-market.png)

Prompt: "A clean product card with the api.market heading and a short tagline"

All other fields were left at their defaults.

### Example Output
The request returned two files. The first is the JSON layer manifest. For this card it reported `resolution` `1344x704`, `is_image_safe` `true`, an empty `image_layers` list and 3 entries in `text_containers`:
- `container_01`: "NEW" (style class `small`, color `#171B3C`, effects `bold` and `all_caps`)
- `container_02`: "api.market" (style class `h1`, font size 75, color `#F7FCFC`) and the tagline "One API key. Hundreds of AI models for images, video, speech and more." (style class `body`)
- `container_03`: "Images", "Video" and "Speech" (style class `small`)

An excerpt of one text item from the manifest:
```json
{
  "alignment": "left",
  "font_file": "PlusJakartaSans-Bold.ttf",
  "font_size": 75,
  "height": 73,
  "spans": [{ "color": "#F7FCFC", "font_fx": [], "text": "api.market" }],
  "style_class": "h1",
  "width": 391,
  "x": 212,
  "y": 205
}
```

The start of `text_html` for the same card:
```html
<div style="position:relative;width:1344px;height:704px;"><div style="position:absolute;left:212.0px;top:121.0px;width:85.0px;height:38.0px;overflow:hidden;...
```

The second file is the background image with the text removed:

![Layerize output: the product card background with all text removed](https://blog.api.market/wp-content/uploads/2026/10/layerize-card-output-api-market.png)

## Input Parameters
| Field | Type | Required | Default | Allowed values | Description |
| --- | --- | --- | --- | --- | --- |
| `flat_graphic_image` | string (URL) | Yes |  |  | The flat graphic image to layerize. Supports JPEG, PNG and WebP formats (max 10MB). |
| `prompt` | string | No | none |  | Text prompt to guide the layerization |
| `seed` | integer | No | none | maximum 2147483647 | Random seed. Set for reproducible generation. |
| `font_name_h1` | string | No | none |  | Font name for heading 1 (h1) text |
| `font_name_h2` | string | No | none |  | Font name for heading 2 (h2) text |
| `font_name_body` | string | No | none |  | Font name for body text |
| `font_name_small` | string | No | none |  | Font name for small text |

## Pricing
- **Pay As You Go (PAYG) pricing**: you only pay for what you use.
- **Price**: $0.115 per API unit.
- **1 API unit = 1 successful request.** Each request returns the JSON layer manifest and the background image together, so the example request costs $0.115.
- **Failed or cancelled predictions are not billed.**
- **GET requests to check prediction status** are free and don't count toward API unit usage.
- **No monthly subscription fee.**

## API Documentation

### Base URL
All API requests go to: `https://prod.api.market/api/v1/magicapi/graphic-text-layer-extraction-ideogram-ai-layerize`

### Authentication
Send your api.market API key in the `x-api-market-key` header:
```
x-api-market-key: YOUR_API_MARKET_KEY
```

### Asynchronous Processing
**IMPORTANT**: This API is asynchronous. The flow is:
1. Send a POST request to `/predictions`. A successful request returns **201 Created** with a prediction `id` and a `status` of `starting` or `processing`.
2. Poll GET `/predictions/{id}` with that ID.
3. Keep polling while `status` is `starting` or `processing`.
4. Stop when `status` is `succeeded` (read `output`), `failed` or `canceled` (read `error`).

A polling interval of 1 second and an overall timeout of 600 seconds (10 minutes) work well.

### Endpoints

#### Create a Prediction
```bash
curl -X POST "https://prod.api.market/api/v1/magicapi/graphic-text-layer-extraction-ideogram-ai-layerize/predictions" \
  -H "Content-Type: application/json" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY" \
  -d '{
    "version": "1586156597a01d63e98fda60f76098963407f251870b6596634420529f226c5d",
    "input": {
      "flat_graphic_image": "https://blog.api.market/wp-content/uploads/2026/10/html-to-image-output-api-market.png",
      "prompt": "A clean product card with the api.market heading and a short tagline"
    }
  }'
```

Response (201 Created):
```json
{
  "id": "bzc0g09c0hrmy0d17sqb6pqkkw",
  "version": "1586156597a01d63e98fda60f76098963407f251870b6596634420529f226c5d",
  "status": "starting",
  "created_at": "2026-10-15T07:19:31.420Z"
}
```

#### Check Prediction Status
```bash
curl -X GET "https://prod.api.market/api/v1/magicapi/graphic-text-layer-extraction-ideogram-ai-layerize/predictions/{id}" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

Response when the prediction has finished (200 OK):
```json
{
  "id": "bzc0g09c0hrmy0d17sqb6pqkkw",
  "status": "succeeded",
  "output": [
    "<URL of the layer manifest .json>",
    "https://blog.api.market/wp-content/uploads/2026/10/layerize-card-output-api-market.png"
  ],
  "error": null,
  "created_at": "2026-10-15T07:19:31.420Z",
  "started_at": "2026-10-15T07:19:32.420Z",
  "completed_at": "2026-10-15T07:19:52.420Z"
}
```

`output` is a list of two URLs. The first points to the JSON layer manifest; the second points to the PNG background with the text removed. The manifest holds a `data` list whose first entry contains:
- `text_containers`: groups of text items. Each item has `spans` (the `text`, its `color` and `font_fx`), `x`, `y`, `width`, `height`, `font_size`, `line_height`, `alignment`, `angle`, `font_file`, `font_candidates` and `style_class` (`h1`, `body`, `small`).
- `text_html`: absolutely positioned HTML of the extracted text, sized to the image.
- `image_layers`: image elements separated from the background (empty for the example card).
- `resolution`: the image size, for example `1344x704`.
- `seed`: the seed used for the request.
- `is_image_safe`: whether the image passed the safety check.

It also repeats the `prompt` and includes `style_type` and `upscaled_resolution`.

### Status Codes
- **201 Created**: prediction created (POST `/predictions`). This is the normal success response.
- **200 OK**: prediction details returned (GET `/predictions/{id}`).
- **400 Bad Request**: invalid input, for example a missing `flat_graphic_image`.
- **401 Unauthorized**: missing or invalid API key.
- **404 Not Found**: no prediction with that ID.
- **429 Too Many Requests**: rate limit exceeded.
- **500 Internal Server Error** or **503 Service Unavailable**: temporary server problem; retry later.

## Rate Limits

All plans allow 10 requests per second. Requests above that return `429 Too Many Requests`; back off before retrying and honour any `Retry-After` header. Polling a prediction with GET is free but still counts toward the per-second limit, so poll every 1 to 5 seconds rather than in a tight loop.

## Error Handling
- Check the HTTP status code of every request and read the response body for the error message.
- A prediction can also end with `status` set to `failed`; the `error` field then explains why. Failed predictions are not billed.
- Retry 429, 500 and 503 responses with exponential backoff. Fix the input before retrying a 400.

## Implementation Guide
### Example Code
Both examples create a prediction, poll until it finishes, then download the layer manifest and print every extracted text item.

#### JavaScript
```javascript
const apiKey = "YOUR_API_MARKET_KEY";
const baseUrl = "https://prod.api.market/api/v1/magicapi/graphic-text-layer-extraction-ideogram-ai-layerize";

async function createAndWaitForPrediction(input) {
  const response = await fetch(`${baseUrl}/predictions`, {
    method: "POST",
    headers: {
      "Content-Type": "application/json",
      "x-api-market-key": apiKey
    },
    body: JSON.stringify({ version: "1586156597a01d63e98fda60f76098963407f251870b6596634420529f226c5d", input })
  });

  if (response.status !== 201) {
    throw new Error(`API error: ${response.status} ${await response.text()}`);
  }

  const prediction = await response.json();
  return pollForResults(prediction.id);
}

const delay = (ms) => new Promise((resolve) => setTimeout(resolve, ms));

async function pollForResults(predictionId) {
  const maxTimeout = 600000;
  const pollInterval = 1000;
  const startTime = Date.now();

  while (Date.now() - startTime < maxTimeout) {
    const response = await fetch(`${baseUrl}/predictions/${predictionId}`, {
      headers: { "x-api-market-key": apiKey }
    });

    if (!response.ok) {
      throw new Error(`API error: ${response.status} ${await response.text()}`);
    }

    const prediction = await response.json();

    if (prediction.status === "succeeded") {
      return prediction.output;
    }

    if (prediction.status === "failed" || prediction.status === "canceled") {
      throw new Error(`Prediction ${prediction.status}: ${prediction.error || "Unknown error"}`);
    }

    await delay(pollInterval);
  }

  throw new Error("Prediction timed out after 10 minutes");
}

async function main() {
  const [manifestUrl, backgroundUrl] = await createAndWaitForPrediction({
    "flat_graphic_image": "https://blog.api.market/wp-content/uploads/2026/10/html-to-image-output-api-market.png",
    "prompt": "A clean product card with the api.market heading and a short tagline"
  });

  console.log("Background without text:", backgroundUrl);

  const manifest = await (await fetch(manifestUrl)).json();
  for (const container of manifest.data[0].text_containers) {
    for (const item of container.items) {
      const text = item.spans.map((span) => span.text).join("");
      console.log(`${item.style_class} at (${item.x}, ${item.y}) size ${item.font_size}: ${text}`);
    }
  }
}

main().catch((error) => console.error(error));
```

#### Python
```python
import time
import requests

API_KEY = "YOUR_API_MARKET_KEY"
BASE_URL = "https://prod.api.market/api/v1/magicapi/graphic-text-layer-extraction-ideogram-ai-layerize"


def create_and_wait_for_prediction(input_data):
    headers = {"Content-Type": "application/json", "x-api-market-key": API_KEY}
    response = requests.post(
        f"{BASE_URL}/predictions",
        headers=headers,
        json={"version": "1586156597a01d63e98fda60f76098963407f251870b6596634420529f226c5d", "input": input_data},
    )

    if response.status_code != 201:
        raise Exception(f"API error: {response.status_code} {response.text}")

    return poll_for_results(response.json()["id"])


def poll_for_results(prediction_id):
    max_timeout = 600
    poll_interval = 1
    headers = {"x-api-market-key": API_KEY}
    start_time = time.time()

    while time.time() - start_time < max_timeout:
        response = requests.get(f"{BASE_URL}/predictions/{prediction_id}", headers=headers)

        if response.status_code != 200:
            raise Exception(f"API error: {response.status_code} {response.text}")

        prediction = response.json()

        if prediction["status"] == "succeeded":
            return prediction["output"]

        if prediction["status"] in ("failed", "canceled"):
            raise Exception(f"Prediction {prediction['status']}: {prediction.get('error') or 'Unknown error'}")

        time.sleep(poll_interval)

    raise Exception("Prediction timed out after 10 minutes")


manifest_url, background_url = create_and_wait_for_prediction({
    "flat_graphic_image": "https://blog.api.market/wp-content/uploads/2026/10/html-to-image-output-api-market.png",
    "prompt": "A clean product card with the api.market heading and a short tagline"
})

print("Background without text:", background_url)

manifest = requests.get(manifest_url).json()
for container in manifest["data"][0]["text_containers"]:
    for item in container["items"]:
        text = "".join(span["text"] for span in item["spans"])
        print(f"{item['style_class']} at ({item['x']}, {item['y']}) size {item['font_size']}: {text}")
```

## FAQ
- **What does one request return?**
  - A list of two URLs: the JSON layer manifest first, then the PNG background with the text removed.
- **How do I put new text back on the image?**
  - Use the background PNG as the base and draw your text at each item's `x`, `y`, `width` and `height`, using its `font_size`, `color` and `alignment`. Or overlay `text_html` on the image and edit the text in the HTML.
- **What kind of images work best?**
  - Flat graphics with clear typography, such as cards, banners and social posts. In one of our tests an illustrated poster returned no text layers, so check `text_containers` before relying on the result.
- **Can I choose the fonts?**
  - Yes. Set `font_name_h1`, `font_name_h2`, `font_name_body` or `font_name_small` to set the font for that text role. Without them, each item reports the font the model matched in `font_file` and alternatives in `font_candidates`.
- **How much does a request cost?**
  - $0.115 per successful request, which returns the manifest and the background image. Failed predictions are not billed.
- **Is the 201 response from POST an error?**
  - No. 201 Created means the prediction was accepted. Use the returned `id` to poll GET `/predictions/{id}` until `status` is `succeeded`, `failed` or `canceled`.

## Contact Us
If you have any questions or need assistance, you can reach out to us through the embedded chat on the bottom right corner of the page or send an email to hello@magicapi.com.


Ideogram Layerize API turns a flat graphic into a clean background plus editable text layers with position, font and color, for redesign and localization.