Unlimited-OCR

• textIgnored

  A text part beside the media is accepted and ignored. A request with no media part is a 400.

• image_urlYes

  One per request. A second image_url part is a 400.

• video_urlNo

  Not accepted.

• document_urlYes

  One PDF, up to 128 pages, rasterized per page at document_dpi (default 96). See Document Inputs.

• Default method markdown; multi_page reads the document through an 8-page sliding window.
• Context window 32,768 tokens, with a maximum output of 24,576 tokens.
• Aliases: baidu/Unlimited-OCR, unlimited-ocr.
• Weights: baidu/Unlimited-OCR.

Both methods return the Markdown-kind payload. multi_page reads an 8-page sliding window per model call by default. Set method_params.window_size to change the window size.

Text mode emits a bare payload string for one image, and one <document> block for a PDF. With markdown, the document contains one <page> block per page of Markdown. The model's own grounding markup is stripped before the reply, so a page reads like any other markdown-kind method. With multi_page, the <document> block carries continuous plain text with no <page> blocks. In JSON mode the reply is the flat document object, with top-level model and method and one record per page under pages. See Methods & Response Format.

from openai import OpenAI

client = OpenAI(
    base_url="https://gateway.vlm.run/v1/openai",
    api_key="<VLMRUN_API_KEY>",
)

response = client.chat.completions.create(
    model="baidu/unlimited-ocr",
    messages=[
        {
            "role": "user",
            "content": [
                {
                    "type": "document_url",
                    "document_url": {
                        "url": "https://storage.googleapis.com/vlm-data-public-prod/hub/examples/finance.sec-filings/tsla-8k.pdf"
                    },
                },
            ],
        }
    ],
    extra_body={"method": "markdown"},
)

print(response.choices[0].message.content)

To read a document in sliding windows, use multi_page and set window_size:

response = client.chat.completions.create(
    model="baidu/unlimited-ocr",
    messages=[
        {
            "role": "user",
            "content": [
                {
                    "type": "document_url",
                    "document_url": {
                        "url": "https://storage.googleapis.com/vlm-data-public-prod/hub/examples/finance.sec-filings/tsla-8k.pdf"
                    },
                }
            ],
        }
    ],
    extra_body={
        "method": "multi_page",
        "method_params": {"window_size": 8},
    },
)