One API, several providers
Azure OpenAI, Kie, and OpenRouter sit behind a single, stable interface.
The brain
The inference service behind the Aether Labs platform. One API on top of several AI providers — built for the latency and reliability that live commentary demands.
Overview
AetherAI sits between the Aether Labs products and the AI providers that power them. It gives the platform a single, stable interface for text generation — and lets the platform choose the right provider for the task, whether that is latency, availability, or cost.
The problem
No single AI provider is the right answer for every match, and managing several of them becomes a job in itself:
Key features
Azure OpenAI, Kie, and OpenRouter sit behind a single, stable interface.
The platform decides the model per call — switching providers does not change the code above it.
Deployed as a Cloudflare Worker — serverless, fast, and close to the network that serves the venue.
Tuned for real-time commentary where seconds matter more than tokens.
Use cases
A match-speaker that announces what is happening, powered by the provider best suited to the moment.
Lineups, rule-of-the-day notes, and welcome copy generated ahead of the event — ready to hand to a voice.
Match notes and short recap copy produced after the event without a separate writing tool.
Who it is for
Teams inside Aether Labs and partners who want one inference endpoint instead of several.
Products that need text generation with real-time pacing and honest failover.
Operators who want commentary and announcements without managing provider accounts themselves.
FAQ
Azure OpenAI, Kie, and OpenRouter are configured today. Adding a provider does not change the interface above it.
It is hosted on Cloudflare by default. Reach out if you need a different deployment story.
No. It is one API for the platform team — you call a single endpoint and get text back.
It powers the Aether Labs family today. If you are building live-event software, contact us — that is exactly the conversation we want to have.
If you are building live-event software, we want to hear about it. Tell us what you are trying to generate in real time.