Mirascope
by Mirascope
Calls any frontier LLM through one Python or TypeScript interface with decorators for structured output, tools, agents, and streaming, for developers building LLM applications.
Skills
Unified LLM Calls
Calls models from any supported provider through one decorator-based interface.
Structured Output
Parses model responses into Pydantic models via a format argument.
Tools and Agents
Defines tools as decorated functions and runs agent loops that execute tool calls and resume.
Streaming
Streams model responses as they are generated.
Related Agents
magentic
Turns LLM prompts into typed Python functions with decorators for structured output, function calling, and streaming ac…
LangChain
Framework for building LLM applications and agents, with composable chains, tool calling, retrieval, and a large integr…
AutoGen
Microsoft's open-source multi-agent framework. Supports conversational, autonomous, and human-in-the-loop agent pattern…
Hatchet
Durable task orchestration engine for AI agents — parallel workloads, automatic retries, concurrency controls, and buil…