List models

GET

/models

const url = 'https://api.katara.com/v1/models';
const options = {method: 'GET'};

try {
  const response = await fetch(url, options);
  const data = await response.json();
  console.log(data);
} catch (error) {
  console.error(error);
}

curl --request GET \
  --url https://api.katara.com/v1/models

Every model the marketplace can serve, with its capabilities and the lowest live price. pricing is absent while no provider is serving the model; a managed model still answers through the fallback at its published price.

Responses

200

The model list.

application/json

object

object

required

string

Allowed value: list

data

required

Array<object>

object

required

Model id, katara/<name>@<version>.

string

object

required

string

Allowed value: model

created

integer

owned_by

required

string

katara

required

object

contextLimit

required

Context window in tokens.

integer

capabilities

required

object

workload

string

Allowed values: chat embedding

tools

boolean

structuredOutputs

boolean

reasoning

boolean

streaming

boolean

maxOutputTokens

integer

pricing

Lowest live offer, in USDC base units (6 decimals) per million tokens. Absent when no provider is serving.

object

input_price_per_mtok

string

output_price_per_mtok

string

Example

{
  "object": "list",
  "data": [
    {
      "id": "katara/llama-3.1-8b-instruct@1",
      "object": "model",
      "created": 0,
      "owned_by": "katara",
      "katara": {
        "contextLimit": 131072,
        "capabilities": {
          "workload": "chat",
          "tools": false,
          "structuredOutputs": false,
          "reasoning": false,
          "streaming": true,
          "maxOutputTokens": 8192
        },
        "pricing": {
          "input_price_per_mtok": "100000",
          "output_price_per_mtok": "300000"
        }
      }
    }
  ]
}