Currently, interaction with Vivian is done through our API. It's our vision to make Vivian accessible across different platforms catering to user preference which is why our API is compatible with any platform supporting OpenAI integration.
To interact with Vivian using Silly Tavern, follow these instructions:
Set the following settings:
Name | Value |
---|---|
API | Chat Completion |
Chat Completion Source | Custom (OpenAI-compatible) |
Custom Endpoint (Base URL) | https://api.vivian.animusai.co |
Custom API Key | Leave this blank |
Enter a Model ID | Leave this blank |
Available Models | Leave this blank |
Select Additional Parameters
and adjust as shown below:
Include Body Parameters
- temperature: 0.65
- top_k: 80
- top_p: 0.95
- max_tokens: 200
- repetition_penalty: 1.2
Include Request Parameters
- Authorization: Bearer <your_animus_api_key>
An option to manage the max_token
size is available here. Vivian's capacity is a 4k context window, which is set to expand shortly.
For API key, use the key given at the time of subscription to a plan.
API Response Configuration
and uncheck streaming
. Streaming is currently disabled in the API and will be available in a future update.