Heym
A workflow automation tool you host yourself, where you drag boxes onto a canvas to connect AI models, actions, and approval steps — instead of paying a monthly fee to a cloud automation platform.
🔗 Visit Heym
Description
Most workflow-automation tools like Zapier or Make are cloud services: your data flows through someone else's servers, and the bill grows with usage. Heym takes a different approach — you install it on your own infrastructure (a laptop, a company server, a cloud VM you control) and build automations on a visual canvas, connecting AI models, tools, and approval checkpoints without writing code.
Under the hood it supports multi-agent orchestration (several AI agents working together on a task), retrieval-augmented generation (RAG) for grounding answers in your own documents, and the Model Context Protocol (MCP) for plugging in external tools. It includes human-in-the-loop approval steps for anything sensitive, real-time cost tracking so you can see exactly what each run costs in tokens and dollars, and a "portal" feature that turns any workflow into a shareable public chat interface. It deploys via Docker Compose or Kubernetes, and its code is source-available under a license that permits self-hosting and internal use but restricts reselling it as a competing service.
💬 Our review
The short version: if the idea of an AI automation platform sending your internal data through a third-party cloud makes you uneasy, Heym is what that same category of tool looks like when you run it yourself — same visual builder, same AI-agent orchestration, but on hardware you control.
It sits in an unusual spot compared to cloud tools like n8n Cloud or Zapier: no monthly subscription per se, but you're responsible for hosting, updates, and uptime yourself, which is a real tradeoff for teams without the ops capacity to babysit a self-hosted service. Compared to n8n (which also offers self-hosting), Heym leans harder into being AI-native from the ground up — RAG and multi-agent coordination are core building blocks rather than bolted-on nodes. For a data-sensitive team already comfortable running Docker or Kubernetes in production, it's a compelling free alternative; for anyone who just wants automation to work without managing a server, a hosted tool remains the simpler choice.
💰 Pricing
📊 Global score
🤖 AI-enriched data
Gratuit pour usage propre et auto-hébergement (licence MIT + Commons Clause, revente interdite). Licence commerciale et support entreprise disponibles sur devis.
Pros
Auto-hébergeable, données restent sous votre contrôle
Orchestration multi-agents, RAG et MCP natifs, pas des extensions
Suivi des coûts LLM en temps réel et traces d'exécution complètes
Cons
Nécessite de gérer soi-même l'hébergement (Docker/Kubernetes)
Licence Commons Clause : revente en tant que service concurrent interdite
Communauté encore jeune comparée à n8n ou Zapier