CONTACT
Request a quote
Tell us the model and the traffic; we answer within a day.
<24h
TO A DEDICATED ENDPOINT
5
MODELS IN THE CATALOG
$0.07
PER MILLION, BLENDED
OPENAI
COMPATIBLE API
What to include
Four things let us price a workload without a call: the model you want to run, the tokens you push per day, the latency target that matters to you (p95 or p99, in milliseconds), and the region your traffic comes from. If you are between two models, name both and we will quote both.
What happens next
We run your workload shape on Nyx and benchmark it against your current provider: same prompts, same day, streamed. You get the numbers either way. If they hold up, we send a flat quote, and a dedicated endpoint is live within 24 hours of your go-ahead.
Prefer plain email? Write to hello@nyxprovider.com with the same four things and it lands in the same queue.