Methods

Most VLM Run Gateway models expose one or more method values that select a specific operation. Pass method (and optional method_params) at the top level of the request body, or via extra_body in the OpenAI Python SDK. If you omit method, the model’s default is applied.

Request body
OpenAI SDK

{
  "model": "paddleocr/pp-ocrv6",
  "method": "ocr",
  "messages": [ ... ]
}

from openai import OpenAI

client = OpenAI(...)
response = client.chat.completions.create(
    model="paddleocr/pp-ocrv6",
    messages=[{...}],
    extra_body={"method": "ocr"},
)

Every model’s live methods, default_method, and extra_body_help are also available on GET /v1/openai/models.

Visual Question Answering (VQA)

Model	Methods
`qwen/qwen3.5-0.8b`	N/A: Free-form multimodal chat. Does not use `method` or `method_params`; send natural-language prompts in `messages`. Accepts up to 64 `image_url` parts, one `video_url`, or a `document_url` PDF that fans out per-page chat. Tune video sampling with `video_fps`, `video_max_frames`, and `video_resolution` at the top level. See Video Inputs.

Document and Image OCR

Model	Methods
`paddleocr/pp-ocrv6`	`ocr`: full detection + recognition. `method_params.lang` selects the recognition language (`en`, `ch`, `japan`, …); optional `method_params.score_threshold` filters low-confidence lines. (default) `detect`: text detection only. Returns bounding polygons in JSON without recognised text. `markdown`: reading-order markdown page output used for the streaming page-block flow.
`zai-org/glm-ocr`	`ocr`: compact OCR text extraction. (default) `markdown`: reading-order markdown page output used for the streaming page-block flow.
`rednote-hilab/dots.mocr`	`parse_layout`: structured layout parse with text for tables, figures, and reading order. (default) `parse_layout_only`: layout regions and bounding boxes without transcribed text. `ocr`: plain OCR text extraction. `markdown`: reading-order markdown page output used for the streaming page-block flow.

Method Parameters

method_params is an optional object passed alongside method at the top level of the request body, or via extra_body in the OpenAI Python SDK. Keys are model- and method-specific; the tables above call out the parameters each method accepts.

Request body
OpenAI SDK

{
  "model": "paddleocr/pp-ocrv6",
  "method": "ocr",
  "method_params": {
    "lang": "en",
    "score_threshold": 0.5
  },
  "messages": [ ... ]
}

response = client.chat.completions.create(
    model="paddleocr/pp-ocrv6",
    messages=[{...}],
    extra_body={
        "method": "ocr",
        "method_params": {"lang": "en", "score_threshold": 0.5},
    },
)

Method-specific keys

Key	Model	Method(s)	Description
`lang`	`paddleocr/pp-ocrv6`	`ocr`, `detect`	Recognition language (`en`, `ch`, `japan`, …). Default `en`.
`score_threshold`	`paddleocr/pp-ocrv6`	`ocr`	Drop text regions below this recognition confidence (0.0–1.0). Default `0.5`.

Multimodal Inputs

Content part types and document input limits.

Models

Availability status and capability matrix for every model.

Chat Completions

Full request and response schema.

Overview

Guides

API Reference

Misc

Visual Question Answering (VQA)

Document and Image OCR

Method Parameters

Method-specific keys

Multimodal Inputs

Models

Chat Completions

​Visual Question Answering (VQA)

​Document and Image OCR

​Method Parameters

​Method-specific keys

​Related

Multimodal Inputs

Models

Chat Completions

Visual Question Answering (VQA)

Document and Image OCR

Method Parameters

Method-specific keys

Related