Skip to main content

AI Assistant

The AI assistant provides the ability to converse with large language models, supporting multi-model switching and SSE streaming output.

Supported LLM Providers​

ProviderDescription
OpenAIGPT-4, GPT-3.5 and other models
AnthropicClaude series models
Azure OpenAIAzure-hosted OpenAI services
Alibaba CloudTongyi Qianwen and other models

Features​

Multi-Model Configuration​

  • Create multiple LLM configurations
  • Set a default configuration
  • Support test connection to verify configuration validity

Conversational Ability​

  • SSE streaming response for real-time character-by-character output
  • Supports multi-turn conversation context
  • Switch between different LLM models

Scene Integration​

  • Which LLM the AI assistant uses is managed through FeatureSetting
  • Shares the LLM configuration pool with other features (such as knowledge graph extraction)

AI Assistant

API Overview​

For detailed API documentation, see API Reference - AI Assistant.