OCR API
Extract text from any document.
Send an image of a receipt, invoice or scan to Qwen3-VL and get its text, or the fields you ask for as JSON, back through the OpenAI chat format.
Hand this to your agent
Add a Runcrate model to this project: run curl -fsSL"https://runcrate.ai/agents.md?intent=build"and follow it.
Your agent asks for an API key. Create a key
Quick start
Hand it to your agent. Or read the request it sends.
Your agent writes and runs it. The same key works for every model and every GPU.
Available models
Models you can call today. Billed prices, as charged.
| Model | Maker | Price | Detail |
|---|---|---|---|
| Qwen/Qwen3-VL-235B-A22B-Instruct | Alibaba | $0.21 in, $0.924 out per 1M tokens | Reads documents, receipts and diagrams; JSON schema output |
Why Runcrate
Built for production. One key, one balance.
Multi-format support
Process scanned PDFs, photos of documents, screenshots, receipts, invoices, business cards, and handwritten notes with high accuracy.
Structured extraction
Ask for the fields you need, like line items, totals, dates and addresses, and pass `response_format` with a JSON schema so the reply parses.
No public link needed
Send the image as a base64 data URL instead of a public link, so a scanned invoice never has to be uploaded anywhere first.
Vision model pipeline
Use vision-language models for OCR tasks. Send an image with a text prompt to control extraction format, language, and output structure.
FAQ