Publications
Read about agentic computing and learn the possible.
Integrate generative AI into your application We are developers building for the future. Our open source agent SDK empowers developers to rapidly build modular, pluggable and trusted agents for genAI apps.
Simple, modular, and scalable: Eidolon is the best choice for your next project.
With Eidolon, agents are services, so there is no extra work when it comes time to deploy; the HTTP server is built in.
Create agents using only specification or customize with your own code. Eidolon provides the power when needed, and the flexibility to customize when desired.
Because agents are services, and with well-defined interfaces, they easily communicate with tools dynamically generated from the OpenAPI JSON schema, as defined by the agent services.
Eidolon is open source and designed to be run in any cloud or on-prem. We are an open source tool built by developers, for developers.
With a focus on modularity, Eidolon makes it easy to swap out components. Grab an off the shelf LLM, RAG impl, tools, etc., or just define your own.
Built-in support for enterprise grade security including multiple authentication methods as well as RBAC support for every agent type. Pluggable interfaces allow developers to choose how their agents are secured.
The Eidolon copilot agent can be used out of the box against any LLM and supports multi-media and RAG using configuration files - all without code.
Use the built-in Eidolon copilot as your human interface to complex agents. The power of Eidolon's agents-as-a-service allows easy integration with any backend agent.
Eidolon publishes configurable React components that can be used in any application. You can use these components as is, customize them to fit your needs, or completely roll your own - you decide.
Easily integrate a retrieval agent into your application. The Eidolon retrieval agent is a pluggable component that can be used with any backend vector store.
Add RAG storage to any application using simple services. Eidolon provides the flexibility to customize RAG to fit your needs.
Easily secure agents using role-based access control (RBAC) ensuring that only the right agents have access to the right data.
Eidolon's modular framework enables easy integration of agents with other agents or into any application. Agents are designed to be easily swapped or upgraded.
Agents are stateless and can be scaled horizontally as needed. You choose how to scale to meet your needs.
Eidolon agents are designed to be run in any cloud or on-prem. You choose where your agents run and how they are deployed.
Step 1: Install Eidolon SDK
Download and install the Eidolon SDK from PYPI or get started with a Template Repository.
Step 2: Create your Agents
Build agent using our out of the box components. Customize them or build new components as needed!
Step 3: Integrate your App
Agents are http services, so you can integrate them into your app with ease or leverage Eidolon's built-in webui to deploy a chatbot interface.
Ready!
Read about agentic computing and learn the possible.
Check out our videos on YouTube to see what is possible with Eidolon.
Come join our community on Discord.
While easy to get started, Eidolon is quite powerful. This page provides documentation on some of the more intricate parts.
FAQs
An Agent is an AI system that goes beyond simple text production. It uses a large language model (LLM) as its central computational engine, allowing it to carry on conversations, do tasks, reason, and display a degree of autonomy. These agents accept prompts in natural language, reason out the necessary tasks, and utilize the LLM to compose a response, rather than simply executing a direct command.
Eidolon is an open-source, service-oriented framework that enables developers to create modular and flexible agents that collaborate and cooperate. It aims to shift the focus from direct LLM interaction to goal-oriented, productivity-enhancing agent development. With a focus on modularity, Eidolon makes it easy to swap out components. Grab an off the shelf LLM, RAG impl, tools, etc., or just define your own.
For generative AI applications, multiple agents are required to solve complex tasks. An Agent Services Framework enables agents to operate both independently and cooperatively across different domains, while providing a consistent development environment for building sophisticated AI-powered tools and resources.
Eidolon facilitates an organized and efficient development process, allowing for easier modification and adaptation as requirements evolve. Ultimately, Eidolon empowers developers to move beyond prompt-and-response interactions and create sophisticated, adaptable AI-powered tools.