Haystack is an open-source AI orchestration framework for building production-ready LLM applications in Python.
Haystack is an open-source AI orchestration framework for building production-ready LLM applications in Python. It lets users design modular pipelines and agent workflows with explicit control over retrieval, routing, memory, and generation. With it you can build scalable RAG systems, multimodal applications, semantic search, question answering, and autonomous agents in a transparent architecture that lets you experiment, customize deeply, and deploy with confidence. It is model- and vendor-agnostic, modular and customizable, and supports an extensible ecosystem of custom components.
Building production-ready LLM applications in Python, including scalable RAG systems, multimodal applications, semantic search, question answering, and autonomous agents, via modular pipelines and agent workflows.
Haystack is an open-source framework for building production-ready RAG pipelines and AI agents. While the entrypoint code shows no direct autonomy signals (subprocess, exec, shell), the README explicitly positions it as a framework for building "Agents" and agentic workflows that can use tools and execute multi-step reasoning loops. As a framework, the library itself does not act, but it is designed to construct agents that call tools and take actions without per-step human confirmation. The "Production Ready RAG & Agents" framing signals it is intended for autonomous agentic deployment.
Side effects
• Tool/function invocation by constructed agents
• Calls to external LLM APIs and model providers
• Reads/writes to document stores and vector databases
• Network requests to retrieval sources and integrated services