> ## 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.

# Toggle thinking on abliteration.ai

> Toggle the thinking step on abliteration.ai with the thinking parameter on Chat Completions and Anthropic Messages.

`abliterated-model` thinks before it answers. Set `"thinking": false` at the top level of the request body to skip the thinking step, or leave it off to keep the default.

## Endpoints

| Endpoint                    | `thinking` toggle |
| --------------------------- | ----------------- |
| `POST /v1/chat/completions` | ✓                 |
| `POST /v1/messages`         | ✓                 |
| `POST /v1/responses`        | Not supported yet |

Default is `true` (thinking on).

## Chat Completions

```sh theme={"system"}
curl -s https://api.abliteration.ai/v1/chat/completions \
  -H "Authorization: Bearer $ABLIT_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "abliterated-model",
    "thinking": false,
    "messages": [{ "role": "user", "content": "Capital of France?" }]
  }'
```

## Anthropic Messages

```sh theme={"system"}
curl -s https://api.abliteration.ai/v1/messages \
  -H "Authorization: Bearer $ABLIT_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "abliterated-model",
    "max_tokens": 256,
    "thinking": false,
    "messages": [{ "role": "user", "content": "Capital of France?" }]
  }'
```