rikkahub/rikkahub
Rikkahub
Rikkahub is an AI model aggregator and frontend interface that provides a unified platform for interacting with multiple large language model providers. It serves as a retrieval-augmented generation chat client with a provider-agnostic gateway, allowing users to switch between different models and endpoints.
The platform features a character persona manager for importing structured character cards and behavior settings to define specific interaction styles. It includes a sandboxed code execution environment with a portable Linux agent for running technical scripts and commands within the chat.
The system manages conversation history through a branch-based tree structure and maintains persistent context across threads via long-term vector memory. It supports multimodal context analysis for images and documents, integrates web search for real-time information retrieval, and employs a rendering pipeline for Markdown and LaTeX.
Features
- AI Aggregators - Acts as a unified platform aggregating multiple AI service providers to allow seamless switching between different models.
- Multi-Model AI Orchestrators - Acts as a unified platform for connecting and orchestrating multiple AI model providers through a single interface.
- Long-term Memory Stores - Implements persistent storage mechanisms to retain historical context across multiple sessions using vector embeddings.
- LLM Chat Interfaces - Ships a web-based conversational interface for interacting with various large language models via compatible APIs.
- AI Execution Sandboxes - Provides a secure, isolated environment specifically designed for AI agents to execute technical scripts and Linux commands.
- AI Provider Integrations - Implements connectors for integrating and switching between multiple external AI model providers seamlessly.
- Persona Configurations - Provides a system for defining AI personalities and behavioral constraints through character cards and settings.
- Memory Persistence - Implements memory persistence allowing AI agents to recall user preferences and past interactions over long periods.
- Persona Management Systems - Offers a comprehensive system for creating and managing AI identities and persona behaviors.
- Provider-Agnostic Model Interfaces - Provides an abstraction layer that standardizes internal requests across multiple LLM providers for seamless model switching.
- Agent Sandboxes - Includes a secure, isolated environment for AI agents to execute technical code and Linux commands.
- Sandboxed Execution Environments - Executes technical tasks within a portable, sandboxed Linux agent environment to ensure host system security.
- Persona Card Imports - Supports importing structured character cards to define specific AI personalities and interaction styles.
- RAG Enabled Clients - Provides a chat interface enhanced with retrieval-augmented generation, web search, and long-term vector memory for persistent context.
- Multimodal Capabilities - Enables the AI to process and analyze images, PDFs, and documents alongside text inputs.
- AI Research Assistants - Integrates web search to retrieve real-time information and cite external sources for grounded AI responses.
- Code Execution Environments - Implements a sandboxed workspace with a portable Linux agent for safely executing agent-generated code within the chat.
- Conversation Branching Systems - Ships a system for creating independent clones of chat sessions to explore alternative response paths.
- Multimodal Analysis Tools - Extracts and analyzes structured data from a combination of visual and textual media for AI processing.
- Multimodal Processing - Integrates multiple data modalities, allowing the AI to analyze images and documents within a shared context.
- Web Search Integrations - Integrates real-time web search capabilities to ground AI responses in current data with external citations.
- Technical Content Renderers - Renders technical content including LaTeX math, markdown, and syntax-highlighted code blocks.
- Markdown and Syntax Highlighting - Provides a rendering pipeline that displays markdown and LaTeX formulas with syntax highlighting for technical responses.
- AI & Machine Learning - AI-powered content discovery and management.
