The Transmission Protocol
Concept:
Every LLM API call is a JSON request with: 'model' (which LLM to use), 'messages' (array of role/content pairs — system sets behavior, user asks the question), 'temperature' (0=deterministic, 1=creative, 2=wild), and 'max_tokens' (response length limit). The system prompt is invisible to the end user but controls everything. This is the OpenAI-compatible format used by Moonshot, Deepseek, and many others.
Science Officer Chen:
Commander, we've been sending ARIA informal messages. But to truly control the intelligence, we need to learn the proper transmission protocol.
Commander Vega:
Protocol? I thought we just send text and get text back.
Science Officer Chen:
That's the surface. Underneath, every transmission is a structured JSON signal. Let me show you the components.
Science Officer Chen:
First: 'model' — which version of ARIA we're addressing. 'kimi-k2.5' is the standard channel. Then: 'messages' — an array of transmissions. Each has a 'role' and 'content'.
Commander Vega:
What roles?
Science Officer Chen:
'system' is the mission briefing — it sets ARIA's behavior. 'You are a navigation assistant. Be precise. Never guess.' ARIA follows this for the entire conversation. 'user' is your actual question. The system prompt is invisible to the end user but controls everything.
Commander Vega:
And temperature? That was in the metadata.
Science Officer Chen:
Temperature controls randomness. 0 means ARIA gives the same answer every time — deterministic, perfect for calculations. 0.7 is balanced. 2.0 is... creative but unreliable. Think of it as 'how much creative freedom do we give the intelligence.' And max_tokens caps the response length — our safety limit.
Commander Vega:
I want to build a proper transmission myself. The full protocol.
Example Code:
{
"model": "kimi-k2.5",
"messages": [
{
"role": "system",
"content": "You are a helpful assistant that explains concepts clearly and concisely."
},
{
"role": "user",
"content": "What is Docker?"
}
],
"temperature": 0.7,
"max_tokens": 200
}Your Assignment
Build a complete API request JSON. Include: 'model' (use 'kimi-k2.5'), 'messages' array with a 'system' role (give it a persona) and a 'user' role (ask a question), 'temperature' (pick a value 0-2), and 'max_tokens' (pick a limit).
Llm Console