OpenAI has significantly expanded ChatGPT's capabilities, benefiting both free and paying users simultaneously. This move reflects how leading AI labs are accelerating value delivery to customers amid intensifying competition.

OpenAI announced on Thursday that it will switch the default model for all users to GPT-5.6 Luna this week. The company stated that compared to the previous default model, GPT-5.5 Instant, Luna reduces the frequency of answers containing at least one factual error by 62%.

Meanwhile, free users and subscribers to the low-cost Go plan will receive unlimited text conversations and a new "Think" button, allowing users to voluntarily activate a more intensive reasoning mode for challenging questions.

This upgrade directly impacts OpenAI's pricing strategy. Just last week, the company reduced prices for both GPT-5.6 Luna and GPT-5.6 Terra. Previously, price cuts typically occurred months after a model's release, but this adjustment has arrived only weeks after launch, indicating a significantly accelerated pace.

Luna Becomes New Default, Free Tier Features Massively Expanded

GPT-5.6 Luna, launched in early July, is positioned as a speed-focused model. OpenAI reports that it offers a substantial improvement in factual accuracy compared to the previous default model, with a 62% reduction in error rate.

In terms of user rights expansion, free users and Go plan subscribers will gain unlimited text conversations along with the newly added "Think" button.

This feature allows users to voluntarily trigger deeper reasoning for complex problems, further raising the intelligence ceiling of the free tier.

Paid Users Receive Sol Model Enhancements, New Reasoning Depth Adjustment Tool

For Plus and Pro paying users, OpenAI has specifically optimized GPT-5.6 Sol to make it more suitable for daily conversations and tasks. This includes quick Q&A, web searches, advice, planning, writing assistance, and decision support.

OpenAI states that the optimized Sol will provide "more focused answers," reducing unnecessary formatting redundancy, maintaining consistent tone, and lowering error rates.

Additionally, paying users will gain a "slider" tool that allows them to manually adjust the depth of thinking ChatGPT applies when generating responses, offering more flexible resource allocation for tasks of varying complexity.

Price Pressure Accelerates Value Delivery, Competition Forces Industry to Speed Up

This upgrade comes against a backdrop of profound changes in the AI competitive landscape. The pricing pressure from low-cost open-source models is pushing OpenAI and Anthropic to accelerate efforts to justify the premium pricing of their high-end models.

OpenAI's price cuts last week for GPT-5.6 Luna and GPT-5.6 Terra broke the previous convention of adjusting prices months after a model's launch. This shift in pace indicates that market pressure has forced top labs to respond quickly with pricing adjustments after product releases.

For investors, the intelligence offered by cutting-edge AI free plans has already surpassed the practical needs of most average users. How enterprise customers will reasonably budget for advanced AI capabilities remains a key question yet to be settled by the industry.