API

Chat completions

An OpenAI-compatible API: change one base URL and keep your code.

PRINET exposes an OpenAI-compatible chat completions API with no logging, billed per token. Existing OpenAI clients work by changing the base URL and API key.

Endpoint

HTTP

POST /v1/chat/completions

Request body

Field	Type	Notes
`model`	string	A model the network serves. List them with GET /v1/models.
`messages`	array	1–100 messages, each with a role (system, user, assistant) and text content.
`stream`	boolean	Stream tokens as server-sent events. Defaults to true.
`temperature`	number	0 to 2.
`max_tokens`	integer	1 to 8192.

Examples

curl

curl $PRINET_API_BASE/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $PRINET_API_KEY" \
  -d '{
    "model": "MODEL_ID",
    "messages": [{ "role": "user", "content": "Hello" }],
    "stream": true
  }'

JavaScript (openai)

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: process.env.PRINET_API_BASE,
  apiKey: process.env.PRINET_API_KEY,
});

const res = await client.chat.completions.create({
  model: "MODEL_ID",
  messages: [{ role: "user", content: "Hello" }],
});

console.log(res.choices[0].message.content);

Python (openai)

from openai import OpenAI
import os

client = OpenAI(
    base_url=os.environ["PRINET_API_BASE"],
    api_key=os.environ["PRINET_API_KEY"],
)

res = client.chat.completions.create(
    model="MODEL_ID",
    messages=[{"role": "user", "content": "Hello"}],
)

print(res.choices[0].message.content)

Listing models

HTTP

GET /v1/models

Billing

Requests are billed per token in $PRINET. Failed jobs are not charged. See Token and payments.