OpenAI removes the daily limit on free ChatGPT chats and upgrades the default model to GPT-5.6 Luna

OpenAI is removing usage limits on text chats for free ChatGPT users, the company announced on Thursday. The change means free users can send unlimited text-only messages without hitting a daily cap. Limits on image generation, file uploads, and voice tools will still apply. The company is also switching the default model for all users from GPT-5.5 Instant to GPT-5.6 Luna, the lightweight member of the GPT-5.6 family that launched in July after government review.

The accuracy improvement is the headline number. OpenAI says Luna produces 62% fewer factual errors than Instant. Free users and $8/month Go subscribers will also get a new Think button that gives the model more time to process harder questions. Plus and Pro subscribers ($20 and $200/month) get a slider tool on GPT-5.6 Sol, the heavyweight model, that lets them dial reasoning effort from high to extra high to pro depending on the task.

OpenAI said it tweaked Sol for everyday tasks like web research, writing, planning, and decision-making. The model should now give more focused answers with less unnecessary formatting. Luna becomes the default this week, while unlimited chats, the Think button, and the Sol slider roll out next week. ChatGPT crossed one billion monthly users in May, and removing limits for free users is a clear play to keep that base growing rather than losing them to Claude, Gemini, or the growing list of free alternatives.

The safety question is worth noting. Lawsuits against OpenAI have alleged that excessive chatbot use has harmed vulnerable users, including minors. Removing usage limits makes that exposure wider. OpenAI says it has added guardrails so models avoid romantic roleplay, age-restricted challenges, and presenting themselves as substitutes for real-world relationships for users under 18. Whether guardrails are enough when the product now has no off switch for text is a question the courts, not the company, will eventually answer.