DigiAI is a versatile platform designed for the creation of AI applications built on LLM Apps. To get started with DigiAI, navigate to [AgentHub] -> [Model Providers] to add and configure the large language models (LLMs) you plan to utilize.
DigiAI supports prominent model providers like the OpenAI GPT series and the Anthropic Claude series. Since each model offers unique capabilities and configurations, it's essential to choose a model that aligns with your application's requirements. Be sure to obtain the API key from the provider's official site before integrating it into DigiAI.
Model Categories in DigiAI
DigiAI organizes its models into five commonly used categories, each serving distinct purposes:
-
System Inference Models: Ideal for tasks such as chat interactions, generating names, and proposing follow-up questions within various applications.
-
Embedding Models: These are utilized for embedding segmented documents within your knowledge base and efficiently processing user queries.
-
Rerank Models: Enhance the search functionalities within LLMs, ensuring more relevant results.
-
Speech-to-Text Models: Transform spoken language into written text, enabling seamless integration in conversational applications.
-
Text-to-Speech Models: Convert written text into spoken language, enabling voice interactions in applications.
As technology evolves and user needs grow, DigiAI is committed to expanding its selection of LLM providers, ensuring that users have access to cutting-edge capabilities and advancements.