A production-grade multi-agent framework in Rust for building, deploying, and coordinating intelligent agents with LLMs.
AutoAgents is a production-grade multi-agent framework written in Rust that enables developers to build, deploy, and coordinate multiple intelligent agents. It provides a modular architecture with type-safe agent models, structured tool calling, configurable memory, and support for both cloud and local LLM backends. The framework solves the problem of creating scalable, performant, and safe AI agent systems for server and edge deployments.
Rust developers and engineers building intelligent multi-agent systems, AI researchers prototyping agentic workflows, and teams needing production-ready, self-hosted AI agent orchestration with local LLM support.
Developers choose AutoAgents for its Rust-based performance and safety guarantees, modular design supporting extensive customization, and unified interface for numerous LLM providers. Its focus on observability, guardrails, and multi-agent coordination makes it suitable for building reliable, scalable agentic applications.
A multi-agent framework written in Rust that enables you to build, deploy, and coordinate multiple intelligent agents
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
The framework is split into separate crates like autoagents-core, autoagents-llm, and autoagents-telemetry, allowing developers to pick and choose components as needed, as detailed in the Components section.
Supports a wide range of cloud and local LLM providers including OpenAI, Anthropic, Ollama, and Llama.cpp with a unified interface, as shown in the Supported LLM Providers table.
Features sandboxed WASM runtime for safe tool execution and integrates OpenTelemetry for tracing and metrics, ensuring reliable and monitorable agent systems, highlighted in the Key Features.
Built in Rust with benchmarks provided, focusing on memory efficiency and concurrent execution, suitable for high-performance server and edge deployments, as emphasized in the Performance section.
Installation requires multiple prerequisites like Rust, Cargo, LeftHook, Python, uv, and maturin, which can be daunting for new users, as listed in the Prerequisites section.
Some LLM providers like Burn and Onnx are marked as experimental, indicating limited stability and support for certain use cases, as noted in the Experimental Providers table.
Requires proficiency in Rust and async programming, with extensive derive macros and configuration, which may be a barrier compared to more accessible Python frameworks.