# Modelux

> The control plane for LLMs. Policy-driven routing, finance-grade budgets, full decision traces, and replay — across every provider, without changing your code.

Modelux is a managed control plane that sits between your application and every major LLM provider (OpenAI, Anthropic, Google, Azure, Bedrock, Groq, Fireworks). You keep your OpenAI-compatible client; Modelux handles routing strategy, cost enforcement, decision logging, audit, and replay. BYO provider keys; flat subscription pricing with no per-token markup.

## Getting Started

- [Overview](https://modelux.ai/docs/index.md): Modelux is the control plane for LLMs. Policy-driven routing, finance-grade budgets, decision traces, and replay across every provider.
- [Quickstart](https://modelux.ai/docs/quickstart.md): Send your first request through Modelux in under 2 minutes.

## Concepts

- [Routing](https://modelux.ai/docs/concepts/routing.md): How routing configs work in Modelux.
- [Providers](https://modelux.ai/docs/concepts/providers.md): Managing provider credentials in Modelux.
- [Projects & API Keys](https://modelux.ai/docs/concepts/projects.md): Projects, API keys, and access control.
- [Budgets](https://modelux.ai/docs/concepts/budgets.md): Spend caps, alerts, and auto-downgrade.
- [Analytics & Logs](https://modelux.ai/docs/concepts/analytics.md): Request logging, analytics, and decision traces.
- [Ensembles](https://modelux.ai/docs/concepts/ensembles.md): Run multiple models in parallel and aggregate their outputs.
- [Caching](https://modelux.ai/docs/concepts/caching.md): Exact-match and semantic caching.
- [Webhooks](https://modelux.ai/docs/concepts/webhooks.md): Event-driven integrations via webhooks.

## Guides

- [Migrating from OpenAI](https://modelux.ai/docs/guides/openai-migration.md): Move your app from direct OpenAI to Modelux in 3 lines.
- [Setting up a fallback chain](https://modelux.ai/docs/guides/fallback-chain.md): Build a reliable routing config with automatic failover.
- [Cost optimization](https://modelux.ai/docs/guides/cost-optimization.md): Strategies to cut your LLM bill with Modelux.
- [A/B testing models](https://modelux.ai/docs/guides/ab-testing.md): Run controlled experiments to compare models in production.
- [Connecting your AI (MCP)](https://modelux.ai/docs/guides/mcp-setup.md): Manage Modelux from Claude Code, Cursor, or any MCP client.
- [SAML SSO overview](https://modelux.ai/docs/guides/sso.md): Connect your identity provider to Modelux so your team signs in with their work account.
- [SSO with Okta](https://modelux.ai/docs/guides/sso-okta.md): Connect Okta as your SAML identity provider for Modelux.
- [SSO with Microsoft Entra ID](https://modelux.ai/docs/guides/sso-entra.md): Connect Microsoft Entra ID (formerly Azure AD) as your SAML identity provider for Modelux.
- [SSO with Google Workspace](https://modelux.ai/docs/guides/sso-google.md): Connect Google Workspace as your SAML identity provider for Modelux.
- [SCIM user provisioning](https://modelux.ai/docs/guides/scim.md): Automate user lifecycle (create, update, deactivate) from your IdP using SCIM 2.0.

## API Reference

- [API overview](https://modelux.ai/docs/api/overview.md): Conventions for the Modelux proxy and management APIs.
- [Chat completions](https://modelux.ai/docs/api/chat-completions.md): POST /v1/chat/completions — the primary inference endpoint.
- [Embeddings](https://modelux.ai/docs/api/embeddings.md): POST /v1/embeddings — vector embeddings for text.
- [Management API overview](https://modelux.ai/docs/api/management.md): REST endpoints for managing Modelux resources.
- [Projects API](https://modelux.ai/docs/api/management-projects.md): Create, list, and manage projects via the Management API.
- [Routing Configs API](https://modelux.ai/docs/api/management-routing.md): Create and manage routing configs, versions, and test them.
- [Providers API](https://modelux.ai/docs/api/management-providers.md): Add, rotate, and monitor provider credentials.
- [Budgets API](https://modelux.ai/docs/api/management-budgets.md): Create and manage spend caps and alerts.
- [Analytics & Logs API](https://modelux.ai/docs/api/management-analytics.md): Query request logs, analytics reports, and decision traces.
- [Webhooks API](https://modelux.ai/docs/api/management-webhooks.md): Manage webhook endpoints and deliveries.

## Optional

- [OpenAPI spec](https://modelux.ai/openapi.yaml): Full management API schema
- [Full docs (concatenated)](https://modelux.ai/llms-full.txt): Every docs page in one file
- [Docs tree (JSON)](https://modelux.ai/docs.json): Programmatic doc discovery
- [Pricing](https://modelux.ai/pricing): Plan tiers and feature comparison
- [Features](https://modelux.ai/features): Detailed feature breakdown