GPT-6 Sol and GPT-6 Luna reasoning models land in the OpenAI API | Wirestack

GPT-6 Sol and GPT-6 Luna reasoning models land in the OpenAI API | Wirestack

What you get

Two new OpenAI reasoning models, GPT-6 Sol and GPT-6 Luna, released on 2026-09-22 for the Responses and Chat Completions APIs.

Prerequisites

An OpenAI API key in OPENAI_API_KEY
The openai npm package (npm install openai)

Steps

Pick a model. Per the changelog, gpt-6-sol costs $2 input / $10 output per 1M tokens and gpt-6-luna costs $0.10 / $0.50. Both take text and image input and return text. Luna has a 1,050,000-token context window and up to 128,000 output tokens.
Change the model in any Responses call. Create luna.mjs:

javascript

import OpenAI from "openai";
const client = new OpenAI();

const response = await client.responses.create({
  model: "gpt-6-luna",
  input: "Write a one-sentence bedtime story about a unicorn.",
});

console.log(response.output_text);

Note: a fix on 2026-09-25 corrected an image-encoding bug that had degraded image understanding in both models, so re-test any vision prompts you tried in the first few days.

Test it

Run node luna.mjs. You should get a one-sentence story; swap in gpt-6-sol to compare quality and cost.