OpenAI Brings Unlimited ChatGPT Text Chats to Free Users

The most significant part of OpenAI’s latest ChatGPT update is not another benchmark score or a new model name. It is the disappearance of one of the biggest restrictions on the free version of ChatGPT.

OpenAI announced on August 6 that Free and Go users will get unlimited text conversations with GPT-5.6 Luna, removing the normal rate limit for everyday text chats. The change begins rolling out next week, while Luna is becoming the default model for Free and Go accounts this week.

For a product that OpenAI now says is used by 1 billion people every week, removing the text-message ceiling is a substantial expansion of free AI access.

The message counter is effectively disappearing

Free ChatGPT accounts have traditionally operated with model usage limits. Once users exhausted their allowance, they could be asked to wait for access to reset or encounter restrictions on certain capabilities.

That changes for ordinary text conversations under the new system.

OpenAI says GPT-5.6 Luna will support unlimited text chats for Free users, meaning users can continue asking questions, drafting content, studying, brainstorming or holding longer conversations without hitting the usual text-based rate limit.

The same change also applies to ChatGPT Go, OpenAI’s lower-priced subscription tier.

There is one qualification: OpenAI says unlimited access remains subject to abuse guardrails, so “unlimited” should not be interpreted as unrestricted automated or abusive usage.

Images, files and other tools still have limits

The new policy does not make every ChatGPT feature unlimited.

OpenAI confirmed that limits will continue to apply to file uploads, images and other tools. TechCrunch reports that separate restrictions also remain around areas such as voice and image generation.

That distinction matters because ChatGPT has expanded far beyond a text chatbot. Free users can already upload files, analyze information, search the web and create images, but these computationally heavier features can still have their own usage limits.

The major change is therefore focused specifically on the part of ChatGPT most people use most frequently: typing a prompt and receiving a text response.

GPT-5.6 Luna becomes the free-tier engine

OpenAI is pairing the unlimited-chat policy with a model upgrade.

GPT-5.6 Luna is replacing GPT-5.5 Instant as the default model for Free and Go users. Luna sits at the efficiency-focused end of OpenAI’s GPT-5.6 family, below Terra and flagship model Sol, but OpenAI has positioned it as a capable model designed to deliver intelligence at much lower computational cost.

The company also claims a sizeable reliability improvement.

In an internal evaluation covering financial, medical and legal questions requiring factual detail, OpenAI said responses containing at least one factual error were 62% less common with GPT-5.6 Luna than with GPT-5.5 Instant. GPT-5.6 Sol reduced those errors by 68%, according to the same evaluation.

Those figures come from OpenAI’s own testing rather than an independent benchmark, but they show the company is pitching Luna as more than simply a cheaper fallback model.

Free users also get a Think button

Unlimited chats are arriving alongside another feature that previously would have felt more closely associated with premium AI access.

Free and Go users will receive a new Think button for questions that require more reasoning. Selecting it gives GPT-5.6 Luna additional time to work through a harder problem before producing its answer.

That creates a two-speed experience on the free plan: quick responses for normal conversations and a deeper reasoning option when the problem requires more work.

OpenAI says both unlimited text conversations and the Think button will begin reaching Free and Go users next week.

Falling AI costs help change the economics

The timing is notable because OpenAI has been aggressively reducing the cost of running its smaller models.

On July 30, just a week before announcing unlimited chats, OpenAI cut the API price of GPT-5.6 Luna by 80%. Luna now costs developers $1 per million input tokens and $6 per million output tokens, compared with $5 and $30 respectively for flagship GPT-5.6 Sol.

OpenAI has not said that the price reduction directly enabled unlimited consumer chats, but the two developments point in the same direction: increasingly capable AI models are becoming cheaper to operate.

At ChatGPT’s scale, those efficiency gains matter.

OpenAI said in February that ChatGPT had reached 900 million weekly active users and 50 million consumer subscribers. Its latest official update now says 1 billion people use ChatGPT each week, meaning the service has added roughly another 100 million weekly users since that February milestone.

Making unrestricted basic text conversations available to a user base of that size would have been significantly harder when frontier AI inference was more expensive.

Paid ChatGPT is moving further up the stack

OpenAI is not removing the reason to pay for ChatGPT altogether.

Plus and Pro users are getting an updated GPT-5.6 Sol, which OpenAI describes as more factual, focused and consistent. Those plans are also receiving a slider that lets users control how much reasoning effort ChatGPT applies to an answer.

That suggests OpenAI is gradually changing the dividing line between free and paid AI.

Basic access to conversational intelligence is becoming more abundant, while subscriptions increasingly differentiate themselves through stronger models, deeper reasoning, larger tool allowances and more computationally intensive capabilities.

For users who mostly open ChatGPT to write emails, summarize text, ask questions, study, brainstorm or get everyday assistance, the free tier just became considerably harder to outgrow.

And for the broader AI market, the move raises the competitive pressure again: the battle is no longer simply over which chatbot is smartest, but how much useful intelligence companies can afford to give away for free.