# Orbitrage

> The convenience layer for LLM apps — one line for routing + observability.

## Docs

- [API keys](https://docs.orbitrage.ai/account/api-keys.md): Mint, scope, and revoke the keys that authenticate your traffic.
- [Plans & billing](https://docs.orbitrage.ai/account/billing.md): Credits, plans, and how spend is metered.
- [Chat Completions](https://docs.orbitrage.ai/api-reference/chat-completions.md): POST /v1/chat/completions — the main, OpenAI-compatible inference endpoint.
- [API overview](https://docs.orbitrage.ai/api-reference/introduction.md): Base URL, authentication, headers, and errors for the Orbitrage gateway.
- [List models](https://docs.orbitrage.ai/api-reference/list-models.md): GET /v1/models — the live model catalog for your account.
- [Responses](https://docs.orbitrage.ai/api-reference/responses.md): POST /v1/responses — the OpenAI Responses API, adapted for routing and tracing.
- [Bring your own key (BYOK)](https://docs.orbitrage.ai/concepts/byok.md): Use your own provider credits for matching models — your account is billed, our pooled credits stay untouched.
- [How it works](https://docs.orbitrage.ai/concepts/how-it-works.md): The architecture end to end — gateway, router engine, telemetry, and dashboard.
- [The Intelligence layer](https://docs.orbitrage.ai/concepts/intelligence.md): Always-on anomaly detection, root-cause trajectories, and on-demand explanations.
- [Models](https://docs.orbitrage.ai/concepts/models.md): The model catalog — tiers, providers, pricing, vision, and context windows.
- [Observability](https://docs.orbitrage.ai/concepts/observability.md): What Orbitrage captures on every call, and how it's attributed to runs, workflows, and end-users.
- [Outage Shield](https://docs.orbitrage.ai/concepts/outage-shield.md): When a provider goes down, Orbitrage detects it across orgs and reroutes to a healthy model — so your requests never hard-fail.
- [Routing](https://docs.orbitrage.ai/concepts/routing.md): How Orbitrage scores every prompt and picks the cheapest model that can handle it.
- [Tools Gateway](https://docs.orbitrage.ai/concepts/tools-gateway.md): Reference a tool by name and Orbitrage runs it server-side with our keys — no tool API keys to manage. You pay per call.
- [Dashboard tour](https://docs.orbitrage.ai/dashboard/overview.md): What each page of the Orbitrage dashboard shows you.
- [Playground](https://docs.orbitrage.ai/dashboard/playground.md): Iterate on prompts and multi-step chains, A/B test against real traffic, and promote a winner.
- [Image generation](https://docs.orbitrage.ai/examples/images.md): Generate and edit images through Orbitrage with the OpenAI Images API — routed, billed, and traced.
- [Per-user attribution](https://docs.orbitrage.ai/examples/per-user-attribution.md): See cost and usage per customer by tagging calls with a user id.
- [Streaming](https://docs.orbitrage.ai/examples/streaming.md): Stream tokens through Orbitrage — the SDK sees a clean OpenAI stream.
- [Tool calling](https://docs.orbitrage.ai/examples/tool-calling.md): A complete agent loop with tools, routed and traced through Orbitrage.
- [Agno](https://docs.orbitrage.ai/integrations/agno.md): Route + trace Agno agents through Orbitrage.
- [Anthropic (Claude)](https://docs.orbitrage.ai/integrations/anthropic.md): Use Claude models through Orbitrage — in plain OpenAI format.
- [CrewAI](https://docs.orbitrage.ai/integrations/crewai.md): Route + trace CrewAI agents through Orbitrage.
- [LangChain](https://docs.orbitrage.ai/integrations/langchain.md): Route + trace LangChain (Python & JS) chat models through Orbitrage.
- [LangGraph](https://docs.orbitrage.ai/integrations/langgraph.md): Use Orbitrage with LangGraph agents.
- [LlamaIndex](https://docs.orbitrage.ai/integrations/llamaindex.md): Route + trace LlamaIndex LLM calls through Orbitrage.
- [OpenAI](https://docs.orbitrage.ai/integrations/openai.md): Route and trace the OpenAI SDK through Orbitrage.
- [Vercel AI SDK](https://docs.orbitrage.ai/integrations/vercel-ai.md): Route + trace the Vercel AI SDK through Orbitrage.
- [Introduction](https://docs.orbitrage.ai/introduction.md): The convenience layer for LLM apps — routing + observability in one line.
- [Alerts & anomalies](https://docs.orbitrage.ai/platform/alerts.md): Webhook and Slack alerts on metric thresholds, plus automatic anomaly detection.
- [Ask Analytics](https://docs.orbitrage.ai/platform/ask-analytics.md): In-app assistant that answers natural-language questions about your telemetry — and creates alerts and charts.
- [MCP server](https://docs.orbitrage.ai/platform/mcp.md): Query your Orbitrage analytics from Claude Desktop, VS Code, Cursor, or any MCP client.
- [Slack](https://docs.orbitrage.ai/platform/slack.md): Run Orbitrage from Slack — onboard, create projects and keys, ask analytics, and get proactive alerts.
- [Quickstart](https://docs.orbitrage.ai/quickstart.md): Route + trace your first LLM call through Orbitrage in under two minutes.
- [Node.js SDK](https://docs.orbitrage.ai/sdks/node.md): Install, initialize, and attribute calls with the Orbitrage Node / TypeScript SDK.
- [Python SDK](https://docs.orbitrage.ai/sdks/python.md): Install, initialize, and attribute calls with the Orbitrage Python SDK.