IX is an open-source platform for designing and deploying autonomous and semi-autonomous LLM-powered agents and workflows. Published on GitHub under the MIT license with about 1,046 stars, it remains actively developed in 2026. IX gives you a visual, no-code way to build agents that can delegate and automate a wide variety of tasks, running in parallel and communicating with each other.
The centerpiece of IX is its no-code agent editor, where you drop and connect nodes into a graph that represents an agent’s cognitive logic. Chat is embedded directly in the editor, so you can test and debug agents rapidly as you build them. Under the hood, the platform is model-flexible: it works with OpenAI models, with experimental support for Anthropic, Google PaLM, and Llama. Agents created with IX can power QA chat bots, generate code, extract and analyze data, augment datasets, and act as research assistants, among other uses.
The platform’s design goals are delegation and scale: rather than a single chatbot, you compose teams of specialized agents that divide work between them. Because the editor visualizes the agent graph, non-specialists can follow how a workflow is put together, while engineers keep control of the underlying models and tools. IX emerged from the first wave of GPT-4 agent frameworks and has continued evolving with the field, making it a still-maintained alternative to code-first frameworks.
IX is completely free and open source under the MIT license. You only pay for the model API usage of whichever provider you connect.
Developers and tinkerers who want to build multi-step, multi-agent LLM workflows visually instead of hand-wiring agent frameworks in code, while keeping full control of the models underneath.