uncensoredlabs
Log inGet started
Docs
API referenceSign in

Get started

Overview

UncensoredLabs is an OpenAI-compatible API for approved uncensored and abliterated models. Connect an application or a compatible coding client using one base URL and an application key. Choose a model from the catalog, fund a prepaid balance, and inspect the token usage and charge for each request.

The base-url swap

Set your client's base URL to https://api.yourdomain.com/v1 and use an UncensoredLabs API key. Chat Completions supports both regular responses and streaming. Your messages keep the familiar OpenAI format.

POST /v1/chat/completions 
curl "https://api.yourdomain.com/v1/chat/completions" \
  -H "Authorization: Bearer $UNCENSOREDLABS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"YOUR_APPROVED_MODEL_ID","messages":[{"role":"user","content":"Hello"}],"max_tokens":256}'

Replace api.yourdomain.com with your deployment's API domain. Choose a model identifier from the catalog.

How it works

Each request names an enabled model. The gateway checks your key, model permissions, limits, and available prepaid credit before forwarding the request to the configured inference provider.

  • Reserve: a conservative maximum is held against your available credit.
  • Verify: provider usage supplies the input, cached-input, and output counts.
  • Settle: the final charge uses the price fixed when the request began. Unused reserved credit is released.

If usage cannot be verified, the request keeps a pending hold for reconciliation. The application stores usage and billing metadata without storing your messages or the model's response.

What's in these docs