JBlanked AI

An OpenAI chat-completion-compatible API, running locally through llama.cpp. It is a drop-in replacement for the OpenAI chat completions endpoint; just point your existing code at the URL below.

Endpoint

https://www.jblanked.com/ai/v1/chat/completions

Fully compatible with the OpenAI chat completions request/response format.

Authentication

Requires an API Key. Pass it as a Bearer token in the Authorization header, same as OpenAI.

Authorization: Bearer YOUR_API_KEY

Current Model

qwen3.5:latest

Allowed Parameters

  • messages
  • temperature
  • top_p
  • max_tokens
  • stop
  • seed
  • presence_penalty
  • frequency_penalty
  • response_format
  • tools
  • stream

Example Request

curl https://www.jblanked.com/ai/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "messages": [
      {"role": "user", "content": "Hello!"}
    ]
  }'
import requests

response = requests.post(
    "https://www.jblanked.com/ai/v1/chat/completions",
    headers={
        "Authorization": "Bearer YOUR_API_KEY",
        "Content-Type": "application/json",
    },
    json={
        "messages": [
            {"role": "user", "content": "Hello!"}
        ],
    },
)

print(response.json())

Privacy

This runs on a local server through llama.cpp. Prompts, responses, and request data are never stored.
Note: This service is free for a limited time.
Support