Documentation
How to install, configure and operate the Ragmux gateway. The pages are generated from the Ragmux repository, where the source files live.
Concept explainers and comparisons live in Learn. Learn
Start
Guides
- ProvidersA model connection is one provider type plus a base URL, an API key and a model name. Projects route chat requests through one connection; RAG stores use one for embeddings.
- Retrieval (RAG)A RAG store is a set of documents embedded with one model connection.
- Users, roles and limitsWho can do what in the management API and dashboard, how logins are protected and audited, and the per-project rate limits, budgets, metrics and retention.
Reference
- ConfigurationRagmux is configured entirely through environment variables.
- REST API referenceThe management API under /admin/api and the OpenAI-compatible client API under /v1: authentication, the role each endpoint needs, examples and the error envelope.
- ChangelogAll notable changes to Ragmux are documented here. The format follows Keep a Changelog and the project uses Semantic Versioning.
Operations
- Backup and restoreRagmux keeps all of its state in one PostgreSQL database plus one secret. Back up both, and you can rebuild a gateway on any machine.
- Security policyRagmux sits between your users and your provider API keys, so we treat reports seriously and fix them quickly. Thank you for helping keep it safe.
- ObservabilityRagmux exports Prometheus metrics at /metrics and OpenTelemetry spans over OTLP/HTTP.
- Running more than one replicaRagmux keeps no durable state in the process.