Uncensored AI for developers: OpenAI-compatible API and CLI

Developers often hit filters in the worst places: a security script, a red-team test, a game dialogue generator, or a data pipeline that touches sensitive text. If your tooling already speaks the OpenAI API, switching models is a two-line change.

Drop-in API

Orrevan exposes an OpenAI-compatible endpoint. Point your client at the base URL and use your key:

from openai import OpenAI

client = OpenAI(base_url="https://orrevan.com/api/v1", api_key="orv_...")
r = client.chat.completions.create(
    model="orrevan/abliterated-model",
    messages=[{"role": "user", "content": "Hi"}],
)
print(r.choices[0].message.content)

Streaming and tool-style integrations work like with other OpenAI-compatible providers. The full reference is in the docs.

Keys you can control

• Per-key limits so a leaked key cannot drain your balance
• Warning mails and easy rotation
• Keys expire after at most 12 months

A CLI for coding work

The Orrevan CLI is a terminal agent that reads and edits your project. Install it with:

npm i -g orrevan
orrevan login
orrevan

It supports different working modes, so you decide how much it does on its own.

Pay for what you use

There is no subscription requirement. You top up credit and each request is billed by tokens, with the price per model shown openly.

A fair warning

Uncensored models are tools. They will not refuse as often, but they still make mistakes, so review generated code before you run it. And not every model behaves the same: test the one you plan to use.

Create an account and generate an API key in a minute.