GPT-6 Luna is the fast, cost-efficient model in OpenAI's GPT-6 family, released in September 2026 together with GPT-6 Sol. OpenAI trained it with methods similar to the flagship Astra and positions it below Sol for workloads where speed and price matter more than depth. In Jantaris, GPT-6 Luna has a 1,050,000-token context window and up to 128K output tokens, which is unusually large for a budget model. OpenAI names chat, classification and lightweight agentic tasks as its typical use.
Best for
- Chatbots and assistants with many users
- Classifying, routing and tagging incoming messages
- Summaries and extraction over long files at low cost
- Lightweight agents that call a few tools per task
How to call it
Change the base URL of the OpenAI SDK and set model="gpt-6-luna":
from openai import OpenAI
client = OpenAI(base_url="https://jantaris.com/v1", api_key="sk-jt-...")
resp = client.chat.completions.create(
model="gpt-6-luna",
messages=[{"role": "user", "content": "Summarize this support thread in 3 bullet points."}],
)
print(resp.choices[0].message.content)FAQ
How cheap is GPT-6 Luna?
Its rate in Jantaris is $0.108 / $0.54 per 1M tokens, one of the lowest in the chat catalog. That is why Luna fits jobs you run thousands of times a day, where a pricier model would multiply the bill.
Can GPT-6 Luna read long documents?
Yes. It has the same 1,050,000-token context window as GPT-6 Sol, so you can pass a long report or a big log file in one request. For hard reasoning on that content, a higher tier will do better.
Should I move from GPT-5.6 Luna to GPT-6 Luna?
OpenAI presents GPT-6 Luna as the newer generation of the same tier. Run your existing prompts against both in Jantaris, compare quality and token usage, then change the model field. Everything else in your code stays the same.
How to call GPT-6 Luna via API
from openai import OpenAI
client = OpenAI(base_url="https://jantaris.com/v1", api_key="sk-jt-...")
resp = client.chat.completions.create(
model="gpt-6-luna",
messages=[{"role": "user", "content": "Write a haiku about Baltic amber."}],
)
print(resp.choices[0].message.content)curl https://jantaris.com/v1/chat/completions \
-H "Authorization: Bearer $JANTARIS_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "gpt-6-luna", "messages": [{"role": "user", "content": "Write a haiku about Baltic amber."}]}'import OpenAI from "openai";
const client = new OpenAI({ baseURL: "https://jantaris.com/v1", apiKey: process.env.JANTARIS_API_KEY });
const resp = await client.chat.completions.create({
model: "gpt-6-luna",
messages: [{ role: "user", content: "Write a haiku about Baltic amber." }],
});
console.log(resp.choices[0].message.content);