Thought Leadership
OpenAI is significantly expanding the capabilities of ChatGPT’s free tier. Users without a subscription will soon receive unlimited text conversations powered by the new default model, GPT-5.6 Luna, along with access to an optional reasoning mode.
At a glance for free users:
- GPT-5.6 Luna becomes the new default model (rolling out this week)
- Unlimited text chats (starting next week)
- New “Think” button for more advanced reasoning
- File uploads, image generation, and tools remain subject to usage limits
- Additional safety measures for younger users
Until now, free users had to deal with message rate limits and a less capable default model. That’s about to change. Over the coming days, GPT-5.6 Luna will replace the current default model for both Free and Go accounts. At the same time, OpenAI plans to eliminate message limits entirely for text-only conversations. Existing quotas will continue to apply to file uploads, image generation, and other advanced tools.
New “Think” Button for More Complex Tasks
For prompts that require deeper analysis — such as planning, research, or other computationally intensive tasks — OpenAI is introducing a dedicated “Think” button. When enabled, Luna spends more time processing a request instead of returning a faster but potentially less accurate response.
The feature is scheduled to launch next week and will include abuse protection. OpenAI says it may restrict excessive use to prevent misuse.
OpenAI Claims Fewer Hallucinations
According to OpenAI, internal evaluations using prompts in finance, healthcare, and legal domains — areas where factual accuracy is especially important — show that GPT-5.6 Luna produced approximately 62% fewer inaccurate responses than GPT-5.5 Instant.
The company notes that this figure comes from its own internal testing and has not been independently verified.
As part of the rollout, OpenAI is also introducing updated safety training for users estimated to be minors. According to the company, the model will avoid romantic role-play, refuse age-inappropriate dare requests, and avoid presenting itself as a substitute for real-world relationships. Additional safeguards will also apply to topics including eating disorders, body image, dangerous activities, and graphic violence. OpenAI says further details are available in a dedicated System Card.
Paid Plans Continue to Offer Additional Features
Subscribers to ChatGPT Plus and Pro will receive an updated version of GPT-5.6 Sol, featuring a new control that lets users adjust the model’s reasoning depth—from fast everyday responses to more deliberate reasoning for planning and research tasks.
OpenAI emphasizes that GPT-5.6 Sol is not part of the free-tier update and remains exclusive to paid subscriptions.
OpenAI Frames the Update as an Access Initiative
Rather than focusing solely on technical improvements, OpenAI presents the announcement as a broader effort to expand access to AI. In its announcement, the company states that “Access shapes opportunity.” It positions the update as another step toward delivering what it describes as “abundant intelligence” to as many people as possible.
(Editorial Team)
