OpenAI GPT 6 Luna
OpenAI's fast, low-cost GPT-6 tier for high-volume chat, classification, and light agentic work, with vision.
GPT 6 Luna is a cloud text model built by OpenAI. It is multimodal: alongside a text prompt it accepts images, then replies with generated text. Its context window handles up to 1.05M input tokens and up to 128K output tokens, and it supports adjustable reasoning effort for harder problems. Generation can be tuned with system instructions. It runs through OpenRouter and OpenAI using your own API key, from $0.10 per million input tokens.
- Released
- this month (Sep 2026)
- Pricing
- $0.10 / 1M in · $0.50 / 1M out
- Inputs
- Text, Images
- Context window
- 1.05M in · 128K out
- Reasoning
- Adjustable effort
- Controls
- System prompt
Examples
Generated with GPT 6 Luna via OpenRouter. The same three prompts run against every text model in the catalog, shown verbatim, so the only thing that changes between two models’ answers is the model.
Explain what an API is to someone who has never written code. Do not use a restaurant or waiter analogy. Plain prose only — no headings, bullet points, or markdown formatting. Maximum 50 words.
An API is a set of rules that lets one software program ask another for information or actions. It defines what requests are allowed and how replies are sent, so programs can work together without knowing how each other is built.
A two-person startup is deciding whether to build their own user authentication or use a hosted provider. Give them a clear recommendation, then the single strongest argument against your own recommendation. Plain prose only — no headings, bullet points, or markdown formatting. Maximum 150 words.
Use a hosted authentication provider. For a two-person startup, building and maintaining secure authentication takes time away from the product and creates security work you don’t need to own. The strongest argument against this choice is vendor lock-in: relying on a provider can make a later migration costly and disruptive.
Write a briefing for a non-technical executive explaining why their company's spending on AI is hard to predict, and what they can do about it. Cover at least three distinct causes. Plain prose only — no headings, bullet points, or markdown formatting. Write exactly 205 words.
AI spending is hard to predict because it combines technology costs with new, usage-driven expenses. Services often charge by consumption: each question, document, image, or automated task can trigger fees, while adoption may rise faster than forecasts. A pilot can become costly when thousands of employees use it. Models and tools change quickly. Teams may switch providers, select more capable models, or add retrieval, security, and monitoring. Prices may fall, but better results often encourage heavier use, so savings are uncertain. Costs are scattered across departments. Employees buy subscriptions, developers launch experiments, and vendors bundle AI into existing contracts. This creates hidden spending, duplicate tools, or unexpected commitments. Data preparation, integration, legal review, and training add costs often omitted from initial estimates. Executives can respond without blocking experimentation. Assign a central owner and require teams to register projects, vendors, users, and outcomes. Set spending limits and alerts; track usage by team and task; review invoices regularly. Run time-limited pilots and compare costs, including people and infrastructure, with measurable benefits. Negotiate contracts covering usage rates, caps, data rights, and exit terms. Reserve a contingency for growth and reassess forecasts quarterly. The goal is not a fixed budget, but visibility, choices, and funding tied to demonstrated value.
OpenAI
OpenAI builds GPT, DALL·E, the Sora family, and the open-weight gpt-oss models, and has been a central force behind the modern wave of generative AI.
openai.com ↗