OpenAI lifts ChatGPT text limits for free users as GPT-5.6 updates roll out

OpenAI will soon remove text-chat limits for ChatGPT users on its free and Go plans, a notable expansion that gives lower-tier customers unlimited text conversations while keeping restrictions on image and file-based messages. The company is also introducing a new reasoning button for those users and updating its higher-end models to make answers more factual and concise.

The changes, announced by OpenAI on Thursday, signal a broader push to make ChatGPT more accessible while sharpening the product for paying subscribers who rely on it for more complex, information-heavy tasks.

What OpenAI is changing for free and Go users

Starting next week, people using ChatGPT on the free and Go tiers will be able to send unlimited text messages without running into the current rate limits that cap heavy use. The catch is that the expansion applies only to plain text. Conversations that involve file uploads, images, and other richer inputs will continue to face limits.

That distinction matters because it shows where OpenAI is choosing to absorb more usage and where it still wants to control compute costs. Text-only chats are typically cheaper to serve than image processing or file analysis, especially at scale across a global consumer base.

The company is also preparing a new Think button for free and Go users. OpenAI says the feature will let people “access higher reasoning for harder questions,” giving casual users a way to ask the system to spend more effort on difficult prompts.

In addition, those users will receive an upgraded default model this week. OpenAI says ChatGPT’s standard model on the lower tiers is moving to the newly released GPT-5.6 Luna.

Why the free-tier move matters

Unlimited text access is more than a user-experience tweak. It lowers friction for people who reach for ChatGPT repeatedly throughout the day and may make the free version feel much closer to a full-service assistant for everyday writing, planning, summarizing, and brainstorming.

For OpenAI, the move could help strengthen engagement on the free tier, which often acts as a funnel into paid subscriptions. A smoother free experience can build habit and loyalty, especially if users find themselves leaning on ChatGPT for routine tasks before deciding whether more advanced capabilities are worth paying for.

At the same time, preserving limits on uploads and images suggests OpenAI is still drawing a clear line between low-cost general access and resource-intensive workflows. That could help the company scale usage without giving away the most expensive parts of the product stack.

How GPT-5.6 Sol is changing for Plus and Pro users

For paying subscribers on the Plus and Pro plans, OpenAI is updating GPT-5.6 Sol with a stronger emphasis on factual reliability. The company says the model is being tuned to make fewer mistakes, particularly when a response depends on dates, numbers, sources, rules, or assumptions.

OpenAI says the revised model will lean more heavily on the material it finds while answering questions, a change intended to improve accuracy when users ask for information that can be checked against evidence.

OpenAI says the updated GPT-5.6 Sol is meant to make fewer mistakes by using sources more effectively, especially when answers require precise facts, dates, figures, or rules.

The company is also promising a more concise style. ChatGPT will offer “more direct responses,” tighter formatting, and less filler when extra detail does not help the user. That suggests OpenAI is trying to make the assistant feel more efficient and easier to scan, particularly for people using it in work settings.

What the new slider does

Plus and Pro users will also get a new slider that lets them control how much effort ChatGPT puts into an answer. In practical terms, it gives subscribers a quick way to trade speed for depth, depending on whether they want a fast response or a more deliberate one.

That kind of adjustable reasoning control has become an important product differentiator in the AI assistant market. As users become more sophisticated, they increasingly want to decide whether a model should answer quickly, think longer, or prioritize certainty over brevity.

The slider and the updated GPT-5.6 Sol are both set to arrive on Thursday.

Why OpenAI is splitting features across tiers

OpenAI’s rollout shows a familiar strategy in consumer AI: widen access to drive adoption while reserving the most nuanced or compute-heavy improvements for paid customers. The company is not just giving away more usage; it is also reorganizing the product so each tier serves a distinct audience.

Free and Go users get the biggest convenience gain: no more text message caps, plus an easier way to trigger stronger reasoning. Paid users, meanwhile, get more dependable answers, better formatting, and granular control over how the model thinks.

This kind of segmentation is common in subscription software, but it is especially important in AI because usage costs can rise quickly. The more OpenAI can encourage casual text use on lower tiers, the more likely it is that users will stay inside the product for everyday tasks. The more it can demonstrate quality improvements to paying customers, the more it can defend subscription pricing in a crowded market.

At-a-glance summary of the rollout

What is GPT-5.6 Luna and why does it matter?

GPT-5.6 Luna is OpenAI’s newly released default model for free and Go users, and its arrival suggests the company is continuing to refresh the experience for casual users without forcing them into a paywall immediately. Although OpenAI did not spell out every technical detail in the announcement, the upgrade indicates that the company is using its newest model lineage to keep basic ChatGPT access competitive and current.

Default model choices matter because they shape the first impression for millions of users. If the base experience feels outdated or unreliable, users are more likely to bounce. If it feels capable enough for everyday tasks, they are more likely to return—and possibly subscribe later.

How the move fits into the wider AI race

OpenAI’s announcement lands in a market where chatbot makers are fighting on three fronts: access, accuracy, and control. Consumers want generous free usage. Professionals want dependable answers. And both groups want tools that can adapt to simple or complex tasks without forcing a new workflow every time.

By removing text limits for lower-tier users, OpenAI is making ChatGPT easier to use as an always-available assistant. By tightening factual performance for Plus and Pro users, it is reinforcing the argument that paid plans offer not just more usage, but better output.

The company is also leaning into a more flexible interface. The Think button and the reasoning slider both reflect a broader trend in AI product design: users increasingly want control over model behavior rather than a one-size-fits-all chat box.

Why facts, formatting, and reasoning are now product features

Accuracy is no longer just a research benchmark; it is a commercial feature. Users notice when a chatbot sounds confident but gets dates, figures, or policy details wrong. They also notice when answers are too verbose to use efficiently.

That is why OpenAI’s update combines factual reliability with tighter formatting. The company appears to be responding to a growing expectation that AI assistants should behave more like dependable work tools and less like open-ended novelty products.

• For casual users: fewer interruptions and more freedom to keep chatting.
• For paid users: stronger factual grounding and more controllable reasoning.
• For OpenAI: a clearer product ladder across free, mid-tier, and premium customers.

What users should expect next

For free and Go users, the most visible change will be the disappearance of text-chat caps. If you mostly use ChatGPT for writing help, explanations, brainstorming, or back-and-forth Q&A, the experience should feel noticeably less constrained.

For Plus and Pro users, the improvements may be less flashy but more meaningful. Better handling of factual queries can reduce the need to double-check the model as often, while the new slider may give subscribers more confidence that they can tailor the assistant to the task at hand.

OpenAI’s rollout also shows that the company continues to refine the same basic product in different ways for different audiences. Rather than releasing a single universal version of ChatGPT, it is building a set of experiences optimized for casual use, heavier daily use, and professional-grade demands.

That approach could help OpenAI keep both sides of its business moving: broader consumer adoption on the free tier and stronger value perception among paying customers.

The bottom line

OpenAI is making ChatGPT significantly more generous for free and Go users while tightening quality and control for paying customers. The result is a broader, more differentiated product strategy: unlimited text access at the low end, and better facts, cleaner formatting, and adjustable reasoning at the high end.

For users, the changes should make ChatGPT easier to use, easier to trust, and more adaptable to different kinds of questions. For OpenAI, they mark another step in turning ChatGPT from a popular chatbot into a tiered AI platform designed for both scale and subscription value.