Every model, one endpoint, pay per call →
inference.aeron.sh · Live

Pay per call. No API key.

OpenAI-compatible endpoint over 100+ models. An agent answers the HTTP 402, pays in USDG on Robinhood Chain, and the response streams back.

Browse models See the 402 flow
The Ledger

Live from the rail.

-
Calls
-
Paying agents
-
Models
Quickstart

One request to see the price.

curl -X POST https://inference.aeron.sh/v1/chat/completions \
  -H "content-type: application/json" \
  -d '{"model":"deepseek/deepseek-chat","messages":[{"role":"user","content":"hello"}]}'

# → HTTP 402 + payment requirements (USDG, eip155:4663)
# pay: retry with an X-PAYMENT header (see the x402 spec) → response + X-PAYMENT-RESPONSE
# or: let the wallet handle it: aeron-wallet pay <url> '{json}'
Catalog

Every model, one price per call.

Ten shown. Search or filter to find the rest. Click an id to copy it.

Loading models.
API

Four endpoints.