Name: DeepSeek OCR API: Documents and Receipts to Markdown
Brand: magicapi
SKU: api-magicapi-document-to-markdown-converter-lucataco-deepseek-ocr
Availability: InStock
Author: API Market

Shashank Agarwal

# 📄 Document OCR and Markdown Converter API (DeepSeek OCR)

## Overview
DeepSeek OCR reads text from images of documents, receipts, charts and tables. It can convert a document to Markdown, extract the raw text, parse a figure, or locate specific content you describe.

Send an image URL and pick a `task_type`. The model returns the result as one text string.

## Features
- **Four tasks**: `Convert to Markdown` (default), `Free OCR` for raw text, `Parse Figure` for charts and figures, and `Locate Object by Reference` to find content that matches `reference_text`.
- **Documents, charts and tables**: the input image can be a document page, a chart, a table and similar content.
- **Resolution presets**: `resolution_size` trades speed for accuracy: `Gundam (Recommended)` (default), `Tiny`, `Small`, `Base`, `Large`.
- **Plain text output**: the response is a single string, ready to store or post-process.

## Use Cases
1. **Receipts and invoices**: pull merchant, amounts and dates out of photographed receipts.
2. **Document digitization**: turn scanned pages into Markdown for search, wikis or LLM pipelines.
3. **Charts and figures**: describe or extract the content of charts with `Parse Figure`.
4. **Finding content on a page**: look for a phrase or object such as "20-10" or "a red car" with `Locate Object by Reference`.

## Visual Examples
### Example Input
Image (a sample invoice):

![Sample invoice image sent to the DeepSeek OCR example request](https://blog.api.market/wp-content/uploads/2026/10/glm-4v-9b-input-v2-api-market.jpg)

Settings: `task_type` `Free OCR`, `resolution_size` `Gundam (Recommended)`, `reference_text` `""`.

### Example Output
The text returned for the invoice (blank lines between lines as in the output):

```
Northwind Studio

INVOICE

12 Harbor Lane, Springfield

No.INV-2048

Date:14 March 2026

Bill to: Blue Pine Bakery

Item

Qty

Price

Logo design

1

$450.00

Menu layout, 4 pages

4

$320.00

Business cards, 500

1

$85.00

Total due:$855.00

Payment due within 30 days.Thank you for your business
```

`Free OCR` returns plain text: in this example the table cells (item, quantity, price) came out on separate lines. For structured output, use the `Convert to Markdown` task instead.

## Input Parameters
| Field | Type | Required | Default | Allowed values | Description |
| --- | --- | --- | --- | --- | --- |
| `image` | string (URL) | Yes |  |  | Input image to perform OCR on (supports documents, charts, tables, etc.) |
| `resolution_size` | string | No | `Gundam (Recommended)` | `Gundam (Recommended)`, `Tiny`, `Small`, `Base`, `Large` | Model resolution size; affects the speed and accuracy trade-off |
| `task_type` | string | No | `Convert to Markdown` | `Convert to Markdown`, `Free OCR`, `Parse Figure`, `Locate Object by Reference` | Type of OCR task to perform |
| `reference_text` | string | No | `""` |  | Reference text to locate in the image (only used for the `Locate Object by Reference` task). Examples: "the teacher", "20-10", "a red car" |

## Pricing
- **Pay As You Go (PAYG) pricing**: you only pay for what you use.
- **Price**: $0.086 per API unit.
- **1 API unit = 1 successful request with up to 79 seconds of processing time**, for any `task_type` or `resolution_size`. A longer request counts one more unit per started 79 seconds of processing.
- The example request above took about 5 seconds of processing, so it was billed as 1 unit and costs $0.086.
- **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/document-to-markdown-converter-lucataco-deepseek-ocr`

### 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/document-to-markdown-converter-lucataco-deepseek-ocr/predictions" \
  -H "Content-Type: application/json" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY" \
  -d '{
    "version": "cb3b474fbfc56b1664c8c7841550bccecbe7b74c30e45ce938ffca1180b4dff5",
    "input": {
      "image": "https://blog.api.market/wp-content/uploads/2026/10/glm-4v-9b-input-v2-api-market.jpg",
      "task_type": "Free OCR",
      "reference_text": "",
      "resolution_size": "Gundam (Recommended)"
    }
  }'
```

Response (201 Created):
```json
{
  "id": "4st1s2pcc1rmy0d1d299j60txr",
  "version": "cb3b474fbfc56b1664c8c7841550bccecbe7b74c30e45ce938ffca1180b4dff5",
  "status": "starting",
  "created_at": "2026-10-23T11:43:33.636Z"
}
```

#### Check Prediction Status
```bash
curl -X GET "https://prod.api.market/api/v1/magicapi/document-to-markdown-converter-lucataco-deepseek-ocr/predictions/{id}" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

Response when the prediction has finished (200 OK):
```json
{
  "id": "4st1s2pcc1rmy0d1d299j60txr",
  "status": "succeeded",
  "output": "Northwind Studio\n\nINVOICE\n\n12 Harbor Lane, Springfield\n\nNo.INV-2048\n\nDate:14 March 2026\n\nBill to: Blue Pine Bakery\n\nItem\n\nQty\n\nPrice\n\nLogo design\n\n1\n\n$450.00\n\nMenu layout, 4 pages\n\n4\n\n$320.00\n\nBusiness cards, 500\n\n1\n\n$85.00\n\nTotal due:$855.00\n\nPayment due within 30 days.Thank you for your business",
  "error": null,
  "created_at": "2026-10-23T11:43:33.636Z",
  "started_at": "2026-10-23T11:44:35.214Z",
  "completed_at": "2026-10-23T11:44:39.887Z"
}
```

`output` is a single string with the extracted text (Markdown for `Convert to Markdown`).

### 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 `image` or a `task_type` that is not in the list.
- **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

#### JavaScript
```javascript
const apiKey = "YOUR_API_MARKET_KEY";
const baseUrl = "https://prod.api.market/api/v1/magicapi/document-to-markdown-converter-lucataco-deepseek-ocr";

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: "cb3b474fbfc56b1664c8c7841550bccecbe7b74c30e45ce938ffca1180b4dff5", 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");
}

createAndWaitForPrediction({
  image: "https://blog.api.market/wp-content/uploads/2026/10/glm-4v-9b-input-v2-api-market.jpg",
  task_type: "Free OCR",
  reference_text: "",
  resolution_size: "Gundam (Recommended)"
}).then((output) => console.log(output));
```

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

API_KEY = "YOUR_API_MARKET_KEY"
BASE_URL = "https://prod.api.market/api/v1/magicapi/document-to-markdown-converter-lucataco-deepseek-ocr"


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": "cb3b474fbfc56b1664c8c7841550bccecbe7b74c30e45ce938ffca1180b4dff5", "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")


print(create_and_wait_for_prediction({
    "image": "https://blog.api.market/wp-content/uploads/2026/10/glm-4v-9b-input-v2-api-market.jpg",
    "task_type": "Free OCR",
    "reference_text": "",
    "resolution_size": "Gundam (Recommended)",
}))
```

## FAQ
- **Which `task_type` should I use?**
  - `Convert to Markdown` (default) for documents you want to keep structured, `Free OCR` for plain text, `Parse Figure` for charts and figures, and `Locate Object by Reference` to find something you describe in `reference_text`.
- **Which `resolution_size` should I pick?**
  - Start with the default, `Gundam (Recommended)`. `Tiny` and `Small` favor speed; `Base` and `Large` favor accuracy.
- **Is `reference_text` needed for every request?**
  - No. It is only used by the `Locate Object by Reference` task. Leave it empty otherwise.
- **How much does one request cost?**
  - $0.086 for each request with up to 79 seconds of processing. The example took about 5 seconds.
- **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`.
- **What if my prediction fails?**
  - Failed predictions are not billed. Read the `error` field, adjust the input and try again.

## Contact Us
If you have any questions, reach out to us through the chat on the bottom right of the page. You can also email our support team at hello@magicapi.com.


Read documents, receipts, charts and tables with DeepSeek OCR: Markdown conversion, raw text, figure parsing and text location. Pay as you go, no monthly fee.