Quickstart

--- title: "Quick Start" description: "Aurora AI gateway quick start: run an OpenAI- and Anthropic-compatible LLM gateway, send your first request, and open the admin panel." icon: "rocket" --- ## Run Aurora OSS Aurora is an OpenAI- and Anthropic-compatible AI gateway. The OSS quick start gives you one local /v1 endpoint, provider configuration through environment variables or YAML, managed gateway keys, usage tracking, audit logs, and admin visibility. Enterprise capabilities and Enterprise deployment are provided through Aurora onboarding. ### 1. Choose an edition | Edition | Available features | Start with | | --- | --- | --- | | OSS | Core gateway, providers, dashboard, managed API keys, usage, audit logs, cache, pools, workflows, basic guardrails, failover, metrics | npx, env vars, config/editions/oss.example.yaml, Docker, Docker Compose, or source | | Enterprise | Everything in OSS plus identity, RBAC, tenants, budget enforcement, advanced routing, observability exports, cluster controls, compliance controls | Contact Aurora for access, deployment guidance, and enterprise onboarding | For the full feature split, see Editions. ### 2. Start Aurora Use the method that matches how you deploy. All options expose the same OpenAI- and Anthropic-compatible API on http://your-aurora-host/v1 unless you change PORT or BASE_PATH. <Tabs> <Tab title="NPX"> Available in: OSS. iaurora installs or locates the native Aurora gateway binary, creates a config file, and runs the OSS gateway by default. ``bash npx -y iaurora ` Enterprise artifacts, licenses, and deployment instructions are provided directly by Aurora during onboarding. If you need Enterprise, contact us instead of using OSS install commands. </Tab> <Tab title="OSS: env vars"> Available in: OSS. Use this when you want the fastest local start and do not need a YAML file. `bash AURORA_MASTER_KEY="change-me" \ OPENAI_API_KEY="$OPENAI_API_KEY" \ LOGGING_ENABLED=true \ USAGE_ENABLED=true \ go run ./apps/aurora ` </Tab> <Tab title="OSS: YAML profile"> Available in: OSS. Use this when you want the OSS edition defaults from config/editions/oss.example.yaml. `powershell $env:AURORA_CONFIG_PATH = "config/editions/oss.example.yaml" $env:OPENAI_API_KEY = "<your OpenAI API key>" .\bin\aurora-oss.exe ` From source, use the Make target: `bash OPENAI_API_KEY="$OPENAI_API_KEY" make run-oss ` </Tab> <Tab title="Docker"> Available in: OSS. `bash docker run --rm -p 8080:8080 \ -e AURORA_MASTER_KEY="change-me" \ -e OPENAI_API_KEY="$OPENAI_API_KEY" \ -e LOGGING_ENABLED=true \ -e LOGGING_LOG_BODIES=false \ -e LOG_FORMAT=text \ -e LOGGING_LOG_HEADERS=false \ aurorallm/aurora ` For production, prefer --env-file .env instead of passing secrets directly on the command line. </Tab> <Tab title="Docker Compose"> Available in: OSS. Start only local infrastructure such as Redis, PostgreSQL, MongoDB, Qdrant, and Adminer: `bash docker compose up -d ` Start the app profile too after setting .env, AURORA_CONFIG_HOST_PATH, and the required mounted file paths from docker-compose.yaml: `bash docker compose --profile app up -d ` </Tab> </Tabs> <Info> Enterprise installation, licenses, private artifacts, deployment profiles, and production onboarding are provided directly by Aurora. If you need Enterprise capabilities, contact us and we will provide the correct package, license, and deployment path for your environment. </Info> <Note> Set at least one provider credential or base URL (OPENAI_API_KEY, ANTHROPIC_API_KEY, GEMINI_API_KEY, DEEPSEEK_API_KEY, XAI_API_KEY, GROQ_API_KEY, OPENROUTER_API_KEY, ZAI_API_KEY, AZURE_API_KEY + AZURE_BASE_URL, ORACLE_API_KEY + ORACLE_BASE_URL, OLLAMA_BASE_URL) or Aurora will have no models to route. Aurora also works well with additional OpenAI- and Anthropic-compatible providers out of the box. </Note> ### 3. Send your first request Use curl or the OpenAI and Anthropic SDKs for Python and JavaScript against the same OpenAI- and Anthropic-compatible endpoint: <CodeGroup> `bash curl curl http://your-aurora-host/v1/chat/completions \ -H "Content-Type: application/json" \ -H "Authorization: Bearer change-me" \ -d '{ "model": "<model id returned by /v1/models>", "messages": [{"role": "user", "content": "Say hello in one sentence."}] }' ` `python Python from openai import OpenAI client = OpenAI( base_url="http://your-aurora-host/v1", api_key="change-me", ) completion = client.chat.completions.create( model="<model id returned by /v1/models>", messages=[{"role": "user", "content": "Say hello in one sentence."}], ) print(completion.choices[0].message.content) ` `javascript JavaScript import OpenAI from "openai"; const client = new OpenAI({ baseURL: "http://your-aurora-host/v1", apiKey: "change-me", }); const completion = await client.chat.completions.create({ model: "<model id returned by /v1/models>", messages: [{ role: "user", content: "Say hello in one sentence." }], }); console.log(completion.choices[0].message.content); ` </CodeGroup> ### 4. Open the Admin Panel Open this URL in your browser: http://your-aurora-host/admin/dashboard <Tip> Dashboard UI is enabled by default (ADMIN_UI_ENABLED=true). Admin API endpoints are at /admin/api/v1/* and use the same bearer auth as the main API. </Tip> ### 5. Automate setup from your server Anything the dashboard does is backed by REST endpoints under /admin/api/v1/*. Server-side automation should authenticate with the master key and call those endpoints directly. For endpoint reference see the Admin API section. Example provider bootstrap: `bash curl -X POST http://your-aurora-host/admin/api/v1/providers \ -H "Authorization: Bearer $AURORA_MASTER_KEY" \ -H "Content-Type: application/json" \ -d '{ "name": "openai-primary", "type": "openai", "api_key": "'"$OPENAI_API_KEY"'", "base_url": "", "api_version": "", "models": "" }' ` Use environment variables or YAML for static boot configuration. Use the admin API when your platform needs to provision keys, workflows, or provider records automatically after the gateway is running. ## Verify Models List currently available models: `bash curl -s http://your-aurora-host/v1/models \ -H "Authorization: Bearer change-me" `` Use one of those model IDs in your requests. ## Admin Panel Preview ### Usage Analytics !Aurora AI gateway usage analytics dashboard) ### Audit Logs !Aurora AI gateway audit logs dashboard) ## Next Steps - Choose the right artifact: Editions - Understand response caching: Cache - Configure production settings: Configuration - Add request policies: Guardrails - Connect coding agents: Coding Agents Overview