PERFECTAI / DEVELOPER API

Your connection starts here.

This is the API base address. To generate a response, send a POST request to the chat endpoint below, or enter the base URL in a client that supports OpenAI-style Chat Completions.

Connection details

Base URL

https://api.perfectai.tech/v1

Model: gemma4:e4b
Chat endpoint: POST /v1/chat/completions
Models: GET /v1/models
Authentication: Authorization: Bearer YOUR_API_KEY
Content type: application/json

Use the API key shared privately by the host. Opening a protected endpoint in a browser without that key returns 401 Unauthorized.

8,192tokens in the configured context

2,048maximum output tokens

512default output tokens

Example JSON request

{
  "model": "gemma4:e4b",
  "messages": [{"role": "user", "content": "Hello!"}],
  "max_tokens": 512,
  "stream": false
}

Send this JSON to https://api.perfectai.tech/v1/chat/completions with the headers above. Read the answer from choices[0].message.content.

Text chat and streaming are supported. Input is limited to 12,000 characters across messages and 64 KiB per request. Capacity is one generation at a time, with 30 chat requests per minute shared across callers.