MVP: docs on withohm.dev · chat edge = local :8081 (or your deploy) · api.withohm.dev returns edge-pending until AWS cutover. Status

Pricing

You pay Ohm. We pay OpenAI (and other providers) from our own balance. Those are separate ledgers.

Pay-as-you-go

EventWhat you pay
Cache hitCheap token hit price — served from Redis
Cache missMiss price — includes upstream fetch
Web fetchPer-fetch ingest when fetch_web_context is used

Live numbers: GET /v1/enterprise/skus and your meter: GET /v1/usage.

Savings (estimates only)

GET /v1/usage and GET /v1/savings expose estimated upstream cost avoided from cache hits — not guaranteed savings. Ohm’s invoice and upstream provider costs are separate ledgers.