What Supabase did for databases, neurwerk.base does for AI agents. One integrated architecture for applications, automation, routing, observability, security, identity, model access, data, runtime, and GitOps. Fully open-source and self-hosted on your infrastructure.
Built on battle-tested open-source tooling
neurwerk.base is a curated, open-source architecture organised into seven connected systems. Together they provide the interfaces, control plane, and platform foundation needed to run AI workloads in one environment.
Chat, workflow, and coding interfaces for existing and new AI solutions.
Invoke agents through schedules, webhooks, events, and messages, then execute them in managed runtimes.
Route LLM and MCP traffic across local, private, and frontier models without tying workloads to one provider.
SSO, roles, secrets, network controls, and layered PII detection and protection across the stack.
Operate policies and access from one control surface with traces, usage, cost and value analysis, logs, and alerting.
Durable block and object storage plus product-local data, search, cache, and future agent memory services.
Signed GitOps delivery, certificate trust, and workload lifecycle management on-premise or on European cloud.
Neurwerk OSS 3rd party OSS External deployed / used Planned ★ Supplier license may be required
Every part of the stack comes with production-grade capabilities, pre-integrated and ready to deploy.
SSO, OAuth, JWT, and role-based access for your entire agent fleet, frontends and custom MCP servers. Connect any identity provider. Built on Keycloak.
Real-time token tracking, cost analysis, and usage monitoring per agent and per user. Combined with smart LLM routing across providers, content filters, PII redaction, and jailbreak detection at the proxy layer.
Define, deploy, and connect agents across functions through neurwerk.studio. Use a wide variety of triggers such as email or text to activate your agents.
Client, team, user, and agent-level memory with prompt injection, full-text session archives, reusable skills, and PostgreSQL/pgvector external storage.
Deploy on AWS, GCP, Azure, or bare metal. No vendor lock-in. Full data sovereignty. EU-first by design.
Ollama, OpenAI, Anthropic, Mistral, Llama or Gemini, one unified interface. Switch models with a config change.
Deploy the full stack today. Self-hosted, open-source, and production-ready.
Star on GitHub →Forever fully open-source.