OpenAI Overhauls Product Line: GPT-5 to Launch Within Months, Free Tier Gets Unlimited Access
TechFlow Selected TechFlow Selected
OpenAI Overhauls Product Line: GPT-5 to Launch Within Months, Free Tier Gets Unlimited Access
Free users can use the basic version of GPT-5 indefinitely, while paid tiers unlock higher intelligence levels and more in-depth research sessions through a $20/month Plus subscription and a $200/month Pro subscription.
Navigating Web3 tides with focused insights
Image source: Generated by Wujie AI
On February 13, OpenAI CEO Sam Altman made an important announcement via social media platform X, declaring that the company will completely revise its artificial intelligence product roadmap, canceling the previously planned standalone release of the "o3" model and instead launching GPT-5, which integrates multiple technologies.
Note: GPT-4.5 and GPT-5 will be released in weeks and months respectively
According to Altman, OpenAI's revised roadmap will proceed in two phases: releasing GPT-4.5 (codenamed Orion) within the coming weeks as a transitional product. Though internally regarded as "the last non-chain-of-thought model" with limited improvements, it will still serve to bridge the technological gap during the interim period. The true game-changer, however, will be GPT-5, launching months later.
GPT-5 is expected to end the split between language models and reasoning models by using a dynamic task-processing mechanism that automatically determines when to deliver rapid responses versus initiating deep thinking, while also integrating voice interaction, canvas creation, real-time search, and "deep research" capabilities.
Altman stated that free users will have unlimited access to the basic version of GPT-5, while paid tiers—Plus at $20 per month and Pro at $200 per month—will unlock higher intelligence intensity and increased deep research usage. This strategy lowers the barrier to entry while transforming technical advantages into tiered monetization.
OpenAI's sudden shift in product roadmap stems directly from ChatGPT's descent into a "model maze" caused by rapid iteration—over the past six months, the platform has密集ly released multiple model versions, forcing users to constantly choose among "GPT-4o for general Q&A," "o3-mini for fast reasoning," "GPT-4o mini for lightweight responses," and even a "scheduled-task version of GPT-4o."
Altman admitted, "The uncontrolled pace of product releases was the main cause of this complexity. We realized users shouldn't struggle with model selection—artificial intelligence should just 'work magically when used.'"
Note: OpenAI's current model roadmap
Currently, the ChatGPT Plus subscription interface resembles a technical menu: GPT-4o, positioned for general use, is labeled "suitable for most questions," while the manually activated "scheduled-task version" allows delayed responses; o1 and o3-mini, focused on math and logic, emphasize "advanced reasoning" and "coding ability" respectively; GPT-4 remains available as the "classic model."
This fragmented experience even extends to the product naming system—from the "o series" reasoning models to the "GPT series" language models—requiring users to understand the differences between two distinct technical paths to use the platform efficiently.
Altman bluntly stated on X: "We hate the model selector as much as our users do. Our goal is to return to magical, unified intelligence."
Join TechFlow official community to stay tuned
Telegram:https://t.me/TechFlowDaily
X (Twitter):https://x.com/TechFlowPost
X (Twitter) EN:https://x.com/BlockFlow_News
