A few months ago we opted to use OpenRouter at the company I work for. And moving to it was one of the best decisions we’ve made for our AI implementation.

The reason? Unmatched stability. OpenRouter handles the load balancing across providers so we can just focus on building features that add value for the end users.

We do some live webinars every once in a while where hunderds of event thousands of people click along live in our software and usually the AI feature’s were the features that got hit the hardest and become slow. But not now, now they just do what they’re supposed to do.

OpenAI having some slower and failing requests? It automatically jumps to Azure! Groq down? No worries, Cerebras picks up the requests!

And the coolest part is that it also gives us almost real time insights into our costs, requests and token usage:

This lets us keep an eye on cost and usage.

And recently they added broadcasting of requests, which allows us to gain insights into our LLM usage in tools like PostHog where we then can run evals, which is even better!

This is innovation at the place you don’t expect it: just a router between some providers.

Great work.