Structured extraction API
Refractr turns chaotic payloads into validated output fields a computer can understand, with a single API call and your own JSON schema.
Why Refractr
A custom-trained extraction model, hosted and maintained by us. No prompt chains, no per-seat pricing, no vendor lock-in.
Latency
A custom-trained extraction model, not a general-purpose LLM wrapped in a prompt. Predictable sub-second response times, not multi-second chains.
Schema, your terms
Pass any JSON schema on every call and get a result immediately. No pre-compilation, no warm-up step, no per-schema deployment. Change fields between requests with no penalty.
Not an LLM wrapper
Refractr runs a custom, optimized model hosted by us, not an OpenAI or Anthropic passthrough. One extraction task, done faster and cheaper than general-purpose models.
Use cases
News wires, invoices, contracts, support tickets, research PDFs: anywhere text has fields, Refractr finds them.
{
"ticker": "NVDA",
"move_pct": 12.0,
"revenue_billions": 22.1,
"currency": "USD"
}
{
"invoice_no": "2847",
"vendor": "Acme Corp",
"date": "2026-01-15",
"total_eur": 1249.00
}
{
"order_id": "88420",
"issue": "damaged",
"action": "refund",
"contact": "[email protected]"
}
Pricing
Flat, predictable pay-as-you-go. No tiers, no seats, no minimums. Top up any amount via Stripe.
Common questions
No. Refractr runs a custom-trained, optimized extraction model hosted by us, not a passthrough to OpenAI, Anthropic, or any general-purpose LLM. It's purpose-built for one task: filling a JSON schema from raw text, which is why it's faster than prompt-based approaches.
Fields declared with typed placeholders (like "__date__") are guaranteed to come back as that type or null, never an empty string or a mistyped value. That guarantee covers the shape of the response; extraction is model-based, so validate values against your own rules and enforce required fields in your own schema validator downstream.
During alpha, requests and responses are logged so we can debug failures and improve extraction quality. We'll always tell you before anything changes. Logs are never sold or shared.
One credit per successful extraction: flat, regardless of document or schema size. Failed extractions (schema violations, timeouts) are free. Every account gets 100 free extractions per day during the alpha, refreshed daily. For higher volume, top up any EUR amount via Stripe.
No. Pass the schema on every call. There's no compile step, no warm-up, no per-schema deployment. Change fields between requests with no penalty.
Five lines of code, five minutes to first response.