Alternatives to Semantic Kernel

Microsoft's open-source framework for building conversational and event-driven AI agents in Python and .NET.

Mastra is an open-source TypeScript agent framework for building durable AI agents and workflows that run for days.

Zhipu GLM delivers open-source LLM models, MaaS APIs, and autonomous agents for Chinese enterprises and developers.

Microsoft's open-source, code-first agent framework for stateful, data-intensive conversational tasks.

Open-source, self-hosted AI agent with persistent memory that connects to 20+ chat platforms and learns from every task.

CowAgent is an open-source, self-hosted AI agent that plans tasks, runs tools, and grows its own memory.

Open-source LLM that turns natural language into API calls for tool-using agents

Open-source multi-agent framework for role-based software engineering

Imbue is an open AI lab building coding agent tools that run in parallel and answer to you, not a vendor.

Microsoft's developer framework for building and hosting agentic AI on Azure, with Python, C#, and Go SDKs and a GA Agent Harness runtime.

Open-source Python framework for building inspectable RAG pipelines and production agents, installable via pip.

Google's MIT-licensed open-source framework for building, evaluating, and deploying production AI agents in five languages.

SmolAgents is Hugging Face's barebones Python library for building AI agents that think in code.

Open-source agentic data assistant: your LLM connects to databases, writes SQL and code, and runs analysis in sandboxes

Hermes Desktop is an open-source desktop AI agent with a built-in learning loop that turns your tasks into reusable skills

Open-source AI agent with smart routing that cuts LLM token costs 60-80% per task.

Ralph Loop is an open-source AI agent loop that codes for days in Docker sandboxes and returns commits.

Eino is a Golang LLM application framework by ByteDance for building agents, chains, and graphs in Go.

DeerFlow is an open-source, self-hosted SuperAgent harness for deep research, coding, and content creation.

Open-source agent harness that routes each task to the best model, one command at a time.

Free MIT-licensed Python framework for multi-agent workflows with handoffs, guardrails, sandbox agents, and voice agents.

Agent evaluation and simulation platform that catches hallucinations before they reach production

Durable execution for background jobs, workflows, and AI agents — written in your own code

An integration layer that gives AI agents and apps managed auth plus 10,000+ prebuilt tools across 3,000+ APIs.

GitLab Duo is GitLab's agentic AI layer, adding specialized AI agents, code review, and policy-governed automation directly into DevSecOps workflows.

Pydantic AI is the Python AI SDK for typed agents, realtime voice, image generation, and embeddings — every model a string swap away.

Coder runs self-hosted cloud development environments with governance for AI coding agents and LLM usage.

Paseo is a self-hosted control plane for running AI coding agents like Claude Code and Codex from any device.

The frontend stack for agentic user experience — drop-in chat, generative UI, and shared state for React apps on any AG-UI backend

Agent enablement platform for governing, testing, and optimizing AI agent skills across your dev team.