Astrbot Chatbot

AstrBot is committed to being an open-source, one-stop Agentic chatbot platform and development framework. You can flexibly configure AstrBot as your ideal AI companion, an efficient intelligent customer service assistant, or a professional enterprise knowledge base. With AstrBot, you can easily deploy and develop LLM-powered intelligent chatbots across a wide range of mainstream messaging platforms.

1. Overview of LLM Configuration

4All API supports configuration for chat models, speech-to-text, embedding, and text-to-speech (TTS is not yet adapted for Astrbot and will be added later).

Integration steps:

Add a service provider
In the list of available service providers below, click to refresh the status; green means available

2. Configure the Chat Model

Configure it as shown in the image. One thing to note here is the timeout setting. It’s best to set it fairly high, for example, I set it to 10 minutes. In chat scenarios, sometimes it’s just a simple greeting, while other times it can be much more complex and require thinking. So it’s better to make the timeout more generous to avoid empty replies caused by timeouts.

3. Configure the Speech-to-Text Model

Nothing special here — usually just use whisper-1

4. Configure the Embedding Model

Pay attention to the following two points:

For the embedding dimension, you generally need to enter the model’s default maximum dimension. To check the model’s dimensional range, go to the Model Marketplace page and search for that model.
For embedding models in the gemini-embedding-001 style Gemini API format, 4All API only supports the openai format! So when selecting a service provider, be sure to choose Integrate OpenAI Embedding

4All API - One-stop AI LLM API aggregation platform
Official website: https://4allapi.com
API Base: https://api.4allapi.com