OpenAI is making one of its biggest free-tier changes yet: ChatGPT Free and Go users are getting unlimited text-based chats, removing a restriction that has shaped how casual users experience the product since its earliest days. The update arrives just as ChatGPT has crossed 1 billion weekly users, and it’s paired with a new default model and several other upgrades rolling out across every subscription tier.
Here’s a full breakdown of what’s actually changing, and a look at how this move stacks up against what Google and Anthropic currently offer on their own free plans.
What’s Actually Changing for Free Users
Starting next week, ChatGPT Free and Go users will get unlimited text chats, a significant shift from the message-capped experience free users have dealt with for years. The change is powered by a new model, GPT-5.6 Luna, which will become the default model for both Free and Go tiers, replacing the previous GPT-5.5.
Free and Go users are also getting a new “Think” button, letting them manually select higher reasoning power for more complex questions, a capability that was previously reserved for paid tiers in most AI chatbots. It’s worth noting that unlimited access applies specifically to text-based conversations. OpenAI confirmed that separate limits will remain in place for files, images, voice, and image generation, so the free tier isn’t becoming fully unlimited across every feature, just the core chat experience.
What’s Changing for Plus and Pro Subscribers
Paying users aren’t being left out of this update. ChatGPT Plus and Pro subscribers are getting access to an upgraded GPT-5.6 Sol model, designed specifically for quick, everyday tasks: answering questions, doing web research, offering advice, planning, writing, and helping with decisions. OpenAI says the new model produces more compact and reliable answers compared to its predecessor.
It’s worth noting this is a distinct version of GPT-5.6 Sol from the one used in Codex and OpenAI’s work-focused tools, which remains unchanged for now, so developers relying on the coding-focused version shouldn’t expect these consumer-facing tweaks to affect their workflows.
Plus and Pro users are also getting a new thinking slider, letting them manually adjust how much reasoning effort the model applies to a given question, tuning it up for genuinely complex, multi-step problems or down for quick, simple queries where speed matters more than depth.
The Accuracy Numbers Behind the Update
OpenAI shared some notable internal benchmark results alongside the announcement. According to the company’s own evaluation, compared to the previous GPT-5.5-Instant model, factual errors dropped 62% for GPT-5.6 Luna and 68% for GPT-5.6 Sol. If those figures hold up under independent testing, it would represent a meaningful accuracy improvement, particularly for free-tier users who are now being routed to Luna as their new default experience.
Rollout Timeline
The update is arriving in stages:
- GPT-5.6 Sol for Plus and Pro users — available immediately
- New default model and Think button for Free and Go users — rolling out this week
- Unlimited text chats and the Think button for harder questions — arriving next week
Now Here’s Something New: How This Stacks Up Against Gemini and Claude’s Free Tiers
OpenAI’s move doesn’t happen in a vacuum. It’s a direct response to a competitive landscape where free-tier generosity has increasingly become a genuine battleground between the major AI labs, not just a courtesy feature.
Google’s Gemini has been moving in a similar direction. Rather than fixed daily message caps, Gemini’s free tier now runs on a compute-based allowance system, where usage is weighted by prompt complexity and feature use rather than a simple message count, refreshing on a rolling basis until hitting a weekly ceiling. Google has also pushed its entry-level paid tier down to a notably low price point, positioning itself as the budget-friendly option in the space.
Anthropic’s Claude, by comparison, has historically kept its free tier more modestly scoped, offering solid conversational access but with tighter usage limits than either ChatGPT or Gemini’s free plans, reserving its most capable models and highest usage allowances for paid Pro and Max subscribers.
What’s notable about OpenAI’s announcement is that it’s the most aggressive of the three moves so far specifically on the text chat front. Rather than just adjusting how usage is measured, OpenAI is removing the text-based cap entirely for its free tier, a step neither Google nor Anthropic has fully matched yet on their comparable plans.
Why This Matters: The Real Strategy Behind Going Unlimited
Removing message limits on a free tier isn’t just a generosity play, it’s a calculated business move, and understanding the logic behind it says a lot about where the AI chatbot market is heading in 2026.
User growth is now the primary battleground, not per-user monetization. With ChatGPT already past 1 billion weekly users, OpenAI’s growth strategy increasingly depends on maximizing how deeply people integrate ChatGPT into daily habits, search, planning, writing, decision-making, rather than squeezing more revenue out of free users through friction. A user who hits a message cap and switches to a competitor for the rest of their day is a lost habit-forming opportunity, not just a lost upsell.
Removing limits reduces churn to competitors. Message caps have historically been one of the most common reasons free users bounce between ChatGPT, Gemini, and Claude depending on which one still has “messages left” at a given moment. Eliminating that friction on the text side keeps users inside OpenAI’s ecosystem more consistently, which matters enormously for building the kind of habitual, daily-use behavior that eventually drives paid conversions.
It positions image, voice, and file limits as the real upsell lever going forward. By keeping those features capped while making basic text chat unlimited, OpenAI is effectively signaling where it wants to draw the free-to-paid line going forward, positioning multimodal features, not simple conversation, as the premium differentiator.
The accuracy improvements make “unlimited but free” less risky. Removing usage caps on a less reliable model would have been a questionable move. Pairing the unlimited rollout with a model OpenAI claims is meaningfully more accurate suggests the company is confident enough in Luna’s reliability to let it carry unlimited free-tier traffic without a significant hit to answer quality or reputation.
What This Means for Everyday Users
For casual ChatGPT users, this is a genuinely useful upgrade. No more watching a message counter tick down during a long research session or a multi-step planning conversation, and the new Think button gives free users manual access to a reasoning mode that previously required a paid subscription in most cases.
For anyone comparing AI chatbots for personal or business use, this update also shifts the calculus. If a completely unrestricted text-based experience is your main priority, ChatGPT’s free tier just became noticeably more competitive against Gemini’s free plan and Claude’s free plan, both of which still apply some form of usage ceiling on their no-cost tiers.
The Takeaway
OpenAI’s move to unlimited free text chats is one of the clearer signals yet that the AI chatbot market has entered a new phase, one where raw usage generosity, not just raw model capability, has become a genuine competitive weapon. With ChatGPT past a billion weekly users and rivals racing to match or undercut its free-tier terms, expect Google and Anthropic to face real pressure to respond with their own free-tier adjustments in the months ahead.
For continuing coverage of ChatGPT updates, AI chatbot competition, and the latest software industry news, keep following Tech News Reports for ongoing updates.
