An open-source, AI-native workflow automation platform — a Rust execution engine, 180+ node types, and a visual drag-and-drop canvas. Self-hosted, multi-tenant, Kubernetes-ready.
⭐ Star on GitHub Quick Start Documentation
🎉 v1.5.1 is out — an AI copilot that edits your canvas (streaming, any OpenAI-compatible model), natural-language workflow generation, node-level field assist that writes prompts / regex / code for you, and typed, declarable outputs. Read the release notes →
Trigix is an enterprise-grade, AI-native workflow automation platform. Build, run, and monitor complex workflows visually — connecting AI models, APIs, databases, and SaaS tools with a drag-and-drop canvas editor. It is a powerful self-hosted n8n alternative built for AI-first automation.
Async DAG scheduling, parallel fan-out/fan-in, retries, timeouts, and cancellation.
Built-in LLM nodes, an agent tool-use loop, RAG over pgvector, MCP support — plus an AI copilot that builds and edits workflows for you.
AI models, 100+ SaaS integrations, data transforms, and control flow — extend with the node SDK.
SSO (OIDC), JWT + RBAC, multi-tenant, encrypted secrets, audit log, Kubernetes Helm chart.
Deploy on Kubernetes with the Helm chart (published to GHCR):
helm install trigix oci://ghcr.io/bj-qizhi/charts/trigix --version 0.3.6 \
--namespace trigix --create-namespace
Or run on a single host with Docker Compose:
docker compose -f docker-compose.prod.yml up -d --build
Full instructions in the README.