l15y/wenda

Features

• LLM Gateways - Provides a centralized API gateway to unify communication between various offline model runners and online service providers.
• AI Provider Gateways - Functions as a centralized AI provider gateway that unifies diverse model runners and services through one interface.
• Conversation State Management - Manages conversation state by maintaining and injecting dialogue history to provide context across interactions.
• Conversation Memory Stores - Implements a persistence layer that tracks and retrieves interaction history to maintain context in agentic workflows.
• Language Model Connectivity - Provides interfaces for establishing communication with both offline model runners and online service providers.
• LLM API Integrations - Integrates multiple external and offline language model providers into a single unified interface.
• LLM Workflow Orchestrations - Implements a system for chaining language model calls and processing steps into automated workflows via custom scripts.
• Private AI Deployments - Provides a self-hosted infrastructure for running language models within internal networks to ensure data privacy.
• RAG Context Retrieval - Retrieves relevant document segments from a knowledge base to ground and enhance large language model responses.
• Local Model Deployment - Provides infrastructure for running and managing language models on internal private hardware for data security.
• Conversation History Stores - Ships a persistence layer for storing and retrieving chronological dialogue history to maintain AI context.
• Vector-Store Augmented Generation - Uses vector databases to perform retrieval augmented generation by injecting relevant document snippets into prompts.
• Retrieval Augmentation - Implements retrieval augmentation to ground model responses using external data from vector stores.
• Private Networks - Enables the creation of isolated and secure network environments for self-hosting language models.
• LLM Orchestration - Includes a workflow orchestrator that uses custom scripts and API calls to automate complex conversation flows.
• Knowledge Base Retrieval - Incorporates a retrieval system that augments model responses with external knowledge from vector databases and search engines.
• Model Provider Integrations - Implements a modular system for connecting and configuring multiple different language model providers.
• Workflow Automation APIs - Provides programmatic interfaces to trigger and manage automated application-specific tasks via custom scripts.
• Automation Workflow Orchestrations - Orchestrates custom scripts and external API calls to automate complex sequences of model interactions.
• LLM Hosting - Offers infrastructure and processes for hosting large language models on private hardware for data security.
• LLM API Workflow Steps - Executes LLM API calls as discrete steps within automated workflow pipelines for content generation.
• Custom Workflow Scripts - Allows the automation of complex conversation flows using custom scripts and model configurations.
• LLM Training and Optimization - Platform for local knowledge base Q&A using various LLMs.
• Open Source Models - Provides a platform for managing and querying multiple models.

Star-Verlauf