Skip to main content
The VLM Run Gateway is an OpenAI-compatible API for visual intelligence: OCR, VQA, detection, embeddings, and transcription behind an interface you already know. Point base_url at https://gateway.vlm.run/v1/openai and your existing SDK calls keep working.

​
Why the Gateway

General-purpose model routers are built for text LLMs, so they cover only a thin slice of visual workloads. The Gateway is built for vision models:
  • One API, many models. Every model speaks OpenAI’s own request shapes, on one base_url and one API key. See Models.
  • Document routing built in. Multi-page PDFs are rasterized and fanned out per page for you, with no splitting or stitching.
  • Method and DPI routing. method picks the operation (ocr, detect, markdown, …); document_dpi tunes how much page detail the model sees.
  • Operational signals included. A live model catalog, per-request usage.cost, and x-request-id on every response.
  • Drop-in for agents. The MCP server exposes file-reading tools to any MCP-aware framework.
Treat model, method, and DPI as one tunable unit: start with the cheapest combination that clears your accuracy bar, and pay for more only where it moves the numbers.

​
Next steps

Quickstart

Prerequisites, plus first VQA and document OCR requests.

Flexible Document OCR

Request knobs, page blocks, and streaming for PDFs.

Models

Catalog, capabilities, and model selection.

Methods

Method and method_params reference.

Pricing

Per-token rates, and per-request usage.cost.

MCP Server

Connect any MCP-aware agent (Pydantic AI, LangChain, Mastra, Claude Code).