One API. Every model.
Zero guesswork.
AuraRoute is the routing layer for AI apps. Tell us what you're building — we pick the model, handle failover, and log every request. Use it for coding, agents, research and chat, and run Hermes and OpenClaw on it for free.
Free daily quota · no credit card · OpenAI-compatible
- coding6.8s200
- research5.7s200
- agent7.1s200
- chat5.3s200
- auto9.5s200
Choose a workload, not a model
Describe the job. The router matches it to the best provider — for coding, research, chat or agents.
Auto
defaultLet the platform decide. The router scores the request and picks the best fit.
Coding
Optimized for code generation, debugging and software development.
Research
Longer, more careful answers for analysis and summarising.
Chat
Fast, capable responses for everyday applications.
Agent
Reliable tool calling and agentic workflows.
One request, from your app to the best provider and back
Send model: "auto". The router scores the request, sends it to the best available provider, fails over when one is down, and records what happened.
- Step 1Your app
Send one OpenAI-style request with model “auto”.
- Step 2Auto router
Scores the request for its workload.
- Step 3Best provider
Picks the best available provider — and fails over if it is down.
- Step 4Response
Result returned and logged with the route it took.
Failover built in. If an upstream returns a 503, the request moves to the next best provider — and the request log shows exactly what served it.
Use Hermes and OpenClaw agents free
Create your API key, connect it to Hermes or OpenClaw, and get to work. A free daily quota, no credit card, and every request routed and failed over for you.
- 1Sign in
Continue with Google at AuraRoute. No form, no credit card.
- 2Create your API key
One click in the dashboard. Free to start.
- 3Connect your agent
Paste the base URL and key into Hermes or OpenClaw.
- 4Enjoy
Every request is routed, failed over and logged for you.
- Base URL
- Shown in your dashboard
- API key
- az_live_••••••••••••
- Model
- auto
Hermes, OpenClaw and any OpenAI-compatible client take the same three values.
Everything between your code and the providers
One endpoint for agents, chatbots and apps — with the reliability and visibility production needs.
Hermes, OpenClaw and any agent
Point any OpenAI-compatible agent at AuraRoute and run it on your free daily quota. One key, no model to choose.
Higher availability
A 503 from upstream never reaches you. A request that fails before its first token moves on to the next best provider.
Keys and usage controls
Create a key per app. Each one has its own daily quota and rate limits, and every request it makes is logged.
Built for builders
The API is OpenAI-compatible, so an app needs a new base URL and the model name “auto”. Nothing else changes.
Full request log
See what served every request — the route it took, the latency and the result — in the dashboard.
Free first
A bounded free daily quota and no credit card. Start building before you spend a cent.
See what's happening behind every request
Keys, usage and request history in one place — so debugging latency and tracking quota stays boring.
Overview
Tokens used, requests, success rate and average latency at a glance.
API keys
A key per app, each with its own daily quota and rate limits.
Requests
Every request with its workload, routed model, provider, latency and status.
Usage
Token and request trends over 24 hours, 7 days or 30 days.
Questions, answered
What does model “auto” do?
You choose a workload, not a model. The router scores each request for its workload and sends it to the best available provider.
What happens when a provider is down?
A 503 from upstream never reaches you. The request fails over to the next best provider, and the request log records what served it.
Do I need to change my code?
Very little. The API is OpenAI-compatible, so you point your existing SDK at the AuraRoute base URL and set the model to “auto”.
Is it free to start?
Yes. There is a bounded free API quota and no credit card is required. The daily quota is shown in your dashboard.
Where can I see what happened to a request?
Every request appears in the dashboard request log, with the route it took and the result. Keys and usage live there too.
Can I limit what a key can spend?
Yes. Each key has its own daily quota and rate limits.
From idea to first request in minutes
Create an API key, make your first request with the OpenAI SDK, and ship your AI feature with routing and observability built in.
Free daily quota · no credit card
