inferva

Endpoints

List models

Every model you can route to.

GET/v1/models

Returns auto plus every model id served by an active provider, with the provider name in owned_by and its price in USDG per 1,000 tokens. No key is required.

Response

data[].idstring

Pass this as model in a chat request.

data[].owned_bystring

Provider name, or inferva for auto.

data[].price_per_1k_usdgnumber

Provider price per 1,000 tokens, before discount and protocol fee.

The list changes as providers join and exit. Use auto if you only care about price.
Request
curl https://dapp.inferva.xyz/v1/models
Response · 200
{
  "object": "list",
  "data": [
    { "id": "auto", "object": "model", "owned_by": "inferva", "description": "Routes to the cheapest active provider" },
    { "id": "basalt-70b", "object": "model", "owned_by": "Basalt Inference", "price_per_1k_usdg": 0.0018 },
    { "id": "basalt-8b", "object": "model", "owned_by": "Basalt Inference", "price_per_1k_usdg": 0.0018 },
    { "id": "nw-large", "object": "model", "owned_by": "Northwind Compute", "price_per_1k_usdg": 0.003 },
    ...
  ]
}
Was this page helpful?
PreviousCreate chat completionNext Changelog