GPT-6 Luna

GPT-6 Luna is the GPT-6 model for jobs that repeat all day long, where each call has to cost almost nothing. OpenAI launched it in September 2026 alongside GPT-6 Sol. Its training borrowed techniques from Astra, and OpenAI places it under Sol for traffic where a fast, cheap reply beats a deep one.

Its context window is the same size as Sol's, which is rare for a budget model.

What you can use it for

• Chatbots and assistants serving many users at once
• Deciding which team or queue an incoming message belongs to
• Low-cost summaries and extraction from long files
• Small agents with two or three tools

Settings and limits

• Context window: 1,050,000 tokens; output up to 128K tokens.
• Whole reports or large logs fit in a single call. If the question about them is hard, a stronger tier will answer better.
• Moving from GPT-5.6 Luna means changing only the model field; the rest of your code stays the same.
• OpenAI aims it at conversation, labeling and simple agent work.

Example request

from openai import OpenAI

client = OpenAI(
    base_url="https://kadraris.com/v1",
    api_key="sk-kd-...",
)
resp = client.chat.completions.create(
    model="gpt-6-luna",
    messages=[{"role": "user",
               "content": "Route this message to sales, "
                          "support or billing."}],
)
print(resp.choices[0].message.content)

Replace the three queues with your own and ask for a single word back.

Questions

How low is the GPT-6 Luna rate?

$0.108 / $0.54 per 1M tokens, near the bottom of the Kadraris chat price list. Failed calls cost $0.

Should I upgrade from GPT-5.6 Luna?

Run your current prompts on GPT-5.6 Luna and GPT-6 Luna, compare quality and token use, then switch the model field.

How does it compare with Gemini Flash?

See Gemini 3.8 Flash vs GPT-6 Luna. For harder tasks on the same content, step up to GPT-6 Sol.

GPT-6 Luna in your code

from openai import OpenAI

client = OpenAI(base_url="https://kadraris.com/v1", api_key="sk-kd-...")
resp = client.chat.completions.create(
    model="gpt-6-luna",
    messages=[{"role": "user", "content": "Suggest three opening shots for a short film about a cafe."}],
)
print(resp.choices[0].message.content)

curl https://kadraris.com/v1/chat/completions \
  -H "Authorization: Bearer $KADRARIS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "gpt-6-luna", "messages": [{"role": "user", "content": "Suggest three opening shots for a short film about a cafe."}]}'

import OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://kadraris.com/v1", apiKey: process.env.KADRARIS_API_KEY });
const resp = await client.chat.completions.create({
  model: "gpt-6-luna",
  messages: [{ role: "user", content: "Suggest three opening shots for a short film about a cafe." }],
});
console.log(resp.choices[0].message.content);