Skip to main content
The models endpoint returns the real-time list of LLMs available on the Pinaivu network. Because Pinaivu is a decentralized marketplace, availability is determined by which GPU nodes are currently online and what models they have loaded. A model appears in this list only when at least one node on the network is ready to serve it.

​
Endpoint

GET https://api.pinaivu.com/v1/models

​
Example Request

cURL
curl https://api.pinaivu.com/v1/models \
  -H "Authorization: Bearer sk-pnv-..."

​
Response

​
data
array
A list of model objects representing every LLM currently available on the network.
Example response
{
  "object": "list",
  "data": [
    {
      "id": "llama3.2:1b",
      "object": "model",
      "created": 1712534400
    },
    {
      "id": "llama3.2:3b",
      "object": "model",
      "created": 1712534400
    }
  ]
}
Model availability changes dynamically as GPU nodes join and leave the network. Always query GET /v1/models before relying on a specific model being available, and handle 503 Service Unavailable responses gracefully in case availability drops between your check and your inference request.