AutoAgents and radkit
These two frameworks are competitors, as both aim to provide a multi-agent framework written in Rust for building and coordinating intelligent agents.
About AutoAgents
liquidos-ai/AutoAgents
A multi-agent framework written in Rust that enables you to build, deploy, and coordinate multiple intelligent agents
Supports both cloud (OpenAI, Anthropic, Groq, etc.) and local LLM backends (Ollama, Llama-cpp, Mistral-rs) through a unified provider interface. Features ReAct executors with structured tool calling via derive macros, sandboxed WASM tool execution, configurable memory backends, and built-in observability via OpenTelemetry. Available as both Rust crate and Python package with feature-gated backends for inference hardware optimization.
About radkit
agents-sh/radkit
Rust Agent Development Kit
Provides a unified LLM interface across Anthropic, OpenAI, Gemini, Grok, and DeepSeek with native A2A protocol support for agent-to-agent communication. Leverages Rust's type system for type-safe tool calling, structured outputs via JSON Schema, and multi-turn conversation management through Thread and Event abstractions. Offers optional feature flags for runtime servers, persistent SQLite task stores, and interactive development UI.
Scores updated daily from GitHub, PyPI, and npm data. How scores work