OpenAI is removing text chat limits for all ChatGPT users and rolling out new GPT-5.6 Luna and Sol models with improved reasoning features.
OpenAI is removing limits on text-based conversations for all ChatGPT users, a change that arrives as the platform has crossed the 1 billion weekly user mark. The update introduces new AI models and reasoning tools designed to make everyday interactions faster and more accurate.
Unlimited Text Chats for Free and Go Users
Free and Go tier users will soon be able to chat without hitting text message caps. The new GPT-5.6 Luna model will power this experience, replacing GPT-5.5 as the default model for these tiers. OpenAI noted that separate limits will still apply to files, images, voice, and image generation, so the unlimited access applies specifically to text conversations.
A New Think Button for Complex Questions
Alongside the model switch, Free and Go users are getting a new “Think” button. This lets people manually select higher reasoning power when a question needs deeper analysis, giving users more control over how the AI processes complex requests without switching to a different plan.
Upgrades for Plus and Pro Subscribers
Plus and Pro users are also seeing changes. They now have access to an upgraded GPT-5.6 Sol model built for quick, everyday tasks like:
- Answering questions
- Web research
- Advice and planning
- Writing assistance
- Decision-making support
OpenAI said this version of GPT-5.6 Sol delivers more compact and robust answers. It’s worth noting this is a separate model version from the GPT-5.6 Sol used in Codex and Work, which remains unchanged.
Thinking Slider Puts Users in Control
Plus and Pro subscribers also get a thinking slider, allowing them to adjust how much reasoning effort the model applies to a given task. Users can tune this based on how complex or multi-step their query is, offering more flexibility than a fixed reasoning mode.
Fewer Factual Errors, According to OpenAI
OpenAI shared internal evaluation results comparing the new models to GPT-5.5-Instant. The company reported that factual errors were:
| Model | Reduction in Factual Errors |
|---|---|
| GPT-5.6 Luna | 62% less common |
| GPT-5.6 Sol | 68% less common |
Rollout Timeline
The updated GPT-5.6 Sol model is already available to Plus and Pro users. The remaining changes for Free and Go users are expected to arrive this week. Unlimited text chats and the new Think button for harder questions will follow next week for those tiers.
What This Means for Users
For casual users, removing chat limits could make ChatGPT more useful for everyday tasks without worrying about hitting a cap mid-conversation. For paying subscribers, the added reasoning controls give more flexibility to balance speed and depth depending on the task at hand. The reported drop in factual errors also suggests OpenAI is prioritizing reliability alongside accessibility as competition in the AI chatbot space continues to grow.
As OpenAI continues scaling ChatGPT to a growing global user base, these changes reflect a broader push to make advanced AI reasoning tools available across both free and paid tiers, rather than reserving them exclusively for premium subscribers.
Frequently Asked Questions
No. OpenAI has confirmed that separate limits will continue to apply to files, images, voice, and image generation. Only text-based chats are becoming unlimited.
GPT-5.6 Luna is the new default model for Free and Go users, while GPT-5.6 Sol is an upgraded model for Plus and Pro users, optimized for quick tasks like research, writing, and planning.
Most changes for Free and Go users are expected this week, while unlimited text chats and the new Think button will roll out the following week.
