[view as .md]

modelux Docs

Using an AI agent? Fetch https://modelux.ai/skill/SKILL.md and drop it into your context — it teaches the agent the modelux proxy API, Management API, MCP tools, routing selectors, and both SDKs. For Claude Code / Cursor, install it as a skill: curl -fsSL https://modelux.ai/skill/install.sh | sh.

Welcome. modelux is the control plane for your LLM stack. You point your OpenAI-compatible SDK at modelux and get policy-driven routing across every provider, finance-grade budgets, full decision traces, and replay experiments — without changing your application code.

[ for AI agents ]
The modelux skill is a single markdown file. Agents can read it directly — no install required:
https://modelux.ai/skill/SKILL.md
Or install it as a Claude Code / Cursor skill so it autoloads on relevant tasks:
curl -fsSL https://modelux.ai/skill/install.sh | sh
Details: Install the Claude Code / Cursor skill.

Start here

What modelux does

  • Policy-driven routing. Fallback chains, cost-optimized, latency-optimized, ensembles, A/B tests, cascades, custom rule DSL across OpenAI, Anthropic, Google, Azure, Bedrock, Groq, Fireworks, DeepSeek, xAI, Mistral, Cerebras, Together, Perplexity, Cohere.
  • Finance-grade budgets. Scoped spend caps with auto-downgrade, alerts, and tag-based attribution.
  • Decision-level observability. Every request stores the full routing decision: attempts, reasons, per-attempt timings and costs.
  • Replay & versioning. Configs are versioned with one-click rollback. Replay historical traffic against candidate configs before you ship them.
  • Audit & governance. Audit log, role-based access, SSO/SAML, access review exports for SOC 2.
  • AI-native management. REST API + MCP server — manage everything from your AI agent.

What modelux doesn’t do

  • Prompt management / versioning (use a dedicated tool)
  • Model fine-tuning or hosting (we route to providers)
  • Prompt evaluation (planned, not shipped)