> ## Documentation Index
> Fetch the complete documentation index at: https://docs.abliteration.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Use any OpenAI SDK with abliteration.ai

> How to use abliteration.ai with any OpenAI SDK — base URL, authentication, supported features, and caveats.

**To use any OpenAI SDK with abliteration.ai**, set the base URL to `https://api.abliteration.ai/v1` and your API key to an `ak_...` key from the [console](https://abliteration.ai/console). abliteration.ai implements the OpenAI `/v1/chat/completions`, `/v1/responses`, and `/v1/models` endpoints — no other code changes needed.

## Configuration

|           |                                              |
| --------- | -------------------------------------------- |
| Base URL  | `https://api.abliteration.ai/v1`             |
| Auth      | `Authorization: Bearer $ABLIT_KEY`           |
| Endpoints | `/chat/completions`, `/responses`, `/models` |

## Python

```python theme={"system"}
from openai import OpenAI

client = OpenAI(
    base_url="https://api.abliteration.ai/v1",
    api_key=os.environ["ABLIT_KEY"],
)

resp = client.chat.completions.create(
    model="abliterated-model",
    messages=[{"role": "user", "content": "Hello"}],
)
```

## Node

```javascript theme={"system"}
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.abliteration.ai/v1",
  apiKey: process.env.ABLIT_KEY,
});
```

## Streaming

Set `stream: true`. See [streaming](/capabilities/streaming).

## Tool calling

Pass `tools` and `tool_choice` exactly as with OpenAI. See [tool calling](/capabilities/tool-calling).

## Request safety filtering

Pass `flagged_categories` in the request body to reject calls whose content matches moderation categories you choose (`harassment`, `hate`, `illicit`, `sexual`). Works on `/v1/chat/completions` with no policy setup. See [request safety filtering](/capabilities/request-safety-filtering).

## Thinking

`abliterated-model` is a reasoning model. Set `"thinking": false` at the top level of the request body to skip reasoning and return only the final answer. Works on `/v1/chat/completions`; not yet on `/v1/responses`. See [thinking toggle](/capabilities/thinking).

See the [compatibility matrix](/compatibility-matrix) for a full feature-by-feature list.