cache

cache. inference

Free inference up to your vDIEM share. Yield on the rest.

Hold vDIEM and a slice of the vault's daily Venice allowance is yours to spend through an OpenAI-compatible API. Your quota is your share of the vault, re-snapshotted every UTC midnight.

  1. Connectthe wallet that holds your vDIEM.
  2. Sign onceone gasless signature proves ownership — you get an API key.
  3. Use it anywherepoint any OpenAI-compatible client at the proxy.

Connect the wallet that holds your vDIEM to see today's free quota and issue an API key.

No vDIEM yet? Deposit DIEM in the vault first. Quota activates at the first UTC-midnight snapshot after you hold vDIEM.

How cache. inference works — read the trust model on the architecture page.

Your key is a bearer token: anyone holding it can spend your quota (never your deposit), so treat it like a password. Saving it in this browser is optional — it is encrypted at rest, but re-issuing on any device is always one signature away. Your quota is your pro-rata share of the vault's daily Venice allowance, re-snapshotted from your vDIEM balance at every UTC midnight — the proxy at https://cache-proxy-production.up.railway.app/v1 meters usage against it.