magentic
by Jack Collins
Turns LLM prompts into typed Python functions with decorators for structured output, function calling, and streaming across OpenAI, Anthropic, and Ollama, for Python developers.
Skills
Structured Outputs
Returns Pydantic models and other typed values from prompt-decorated functions based on the return annotation.
Function Calling
Lets the LLM choose Python functions to call and resolves chains of calls automatically with @prompt_chain.
Streaming
Streams text and structured objects as they are generated, including concurrent streams and async variants.
Chat Prompting
Builds prompts from system, user, and assistant message templates for few-shot and multi-turn calls.
Related Agents
Mirascope
Calls any frontier LLM through one Python or TypeScript interface with decorators for structured output, tools, agents,…
AgentScope
Builds multi-agent applications with a Python framework focused on transparency, observability, and tracing. Includes a…
AutoGen
Microsoft's open-source multi-agent framework. Supports conversational, autonomous, and human-in-the-loop agent pattern…
fast-agent
Define and run multi-agent workflows in Python with first-class MCP support — chains, routers, parallel fan-out, evalua…