A Revolution in ChatGPT's Free Tier: Limits on Text Chats Are Going Away

With a radical step that will shift the balance of the AI ecosystem, OpenAI is making an important update to ChatGPT's access policies. According to the company's announcement, the usage limits applied to ChatGPT's Free and Go subscription tiers are being removed completely for text-only chats.

With this change, users will be able to continue their text-based interactions without worrying about a daily message quota. Alongside the unlimited access move, there are also notable changes to the default model architecture and the paid tiers.

The Era of Unlimited Use in Text Chats

Under OpenAI's new decision, no limit will be applied to text-focused conversations for people on the Free and Go plans. Users will get an uninterrupted experience in core use cases such as developing ideas, writing, Q&A and brainstorming.

This flexibility is not limited to the standard response process. The unlimited use of text chats will also remain in effect when the new "Thinking" mode added to the user interface is turned on.

The Default Model Is Changing: GPT-5.6 Luna and Unlimited Thinking

At the same time as the access limits are relaxed, the default AI model for Free and Go accounts is being renewed. The previously offered GPT-5.5 Instant model is giving way to GPT-5.6 Luna, the lightweight and fast member of OpenAI's new-generation GPT-5.6 model family.

The "Thinking" button that comes with GPT-5.6 Luna lets the model take extra reasoning steps before answering complex questions. When this option is selected, the model analyzes the topic in depth and produces more comprehensive and consistent answers. This thinking capability offered within text chats can also be used without limits.

Limits Remain for Images, Files and Voice Mode

While limits on text-based messaging are going away, the existing quota policies will continue to apply to additional features that require heavy compute and data transfer.

Users are said to be subject to the relevant usage limits in the following cases:

• File and Document Uploads: When PDFs, code files or data tables are included in the analysis,
• Image Uploads and Analysis: When an image is added to the chat or a question is asked about an image,
• Image Generation: When new images are generated using the ChatGPT Images / DALL-E infrastructure,
• Advanced Voice Mode: When real-time voice interaction features are used.

When such multimodal operations take place, the system will apply the standard usage limits.

New Features for Plus and Pro Subscribers: GPT-5.6 Sol and the Thinking Slider

OpenAI is also offering improvements that will enhance the platform experience for its paid subscribers. The company's flagship model, GPT-5.6 Sol, is being given a more optimized structure for everyday conversations and complex workflows.

The updated GPT-5.6 Sol model includes the following improvements:

• Dynamic Response Length: Automatically adjusts the length and depth of answers depending on how detailed the question is.
• Simple Formatting: Delivers information directly, avoiding unnecessary list formatting or overly ornate wording.
• Constructive Corrections: Instead of blindly agreeing with the user's input, it identifies incorrect or missing points and gives constructive feedback.

In addition, thanks to the new thinking level slider added to the Plus and Pro user interface, users will be able to manually control how deeply the model reasons about a question.

Why Is OpenAI Offering Unlimited Access Now?

Considering the server and infrastructure running costs of AI models, OpenAI's move to remove text limits points to technological progress. Although the company has not shared official financial or technical details, the significant drops recently achieved in model inference costs are believed to have been decisive in this decision.

OpenAI's official statement stresses that the main goal is to make advanced AI tools accessible to a wider audience. With unlimited text chats, the aim is for users to benefit from AI without worrying about being cut off midway.