Lithune Logo
Checking service

Blazing Fast
AI Inference

Get Started on OpenRouter Current pricing and availability are shown on OpenRouter.
Context length
Maximum output
Infrastructure · Private hardware
API Quickstart

OpenAI-compatible

● ● ●Python · OpenAI SDK
from openai import OpenAI

client = OpenAI(
    base_url=,
    api_key="sk-or-...",
)

response = client.chat.completions.create(
    model=,
    messages=[{"role": "user", "content": "Hello!"}],
    stream=True,
)

for chunk in response:
    print(chunk.choices[0].delta.content or "", end="")

Model inference runs on Lithune-owned hardware in Türkiye. Requests may be routed through OpenRouter and Cloudflare. Lithune does not forward prompts to an external model or GPU provider and does not retain prompt or completion content.