TLDR

• OpenAI plans to launch GPT-5 in the coming weeks or months, following the release of GPT-4.5 (Orion)
• The company will merge all its AI models into a single unified system that includes text, image, and video capabilities
• GPT-5 will be available through ChatGPT’s free tier with standard intelligence settings
• Plus and Pro subscribers will get access to higher intelligence settings
• The move comes after Chinese competitor DeepSeek launched a simpler, more capable AI model

OpenAI CEO Sam Altman has announced plans to streamline the company’s product lineup by combining its various AI models into a single unified system. The announcement comes as the company prepares to launch GPT-5 in the coming weeks or months.

The current OpenAI product line includes several different models: GPT for creative tasks, the “o” series for reasoning, Dall-e for image generation, Sora for video creation, and GPT-Vision for visual understanding. This variety has led to what some users find to be a confusing experience.

“We want to do a better job of sharing our intended roadmap and a much better job simplifying our product offerings,” Altman wrote on social media. He added that the company wants AI to “just work” for users, acknowledging that their current model offerings have become too complicated.

OPENAI ROADMAP UPDATE FOR GPT-4.5 and GPT-5:

We want to do a better job of sharing our intended roadmap, and a much better job simplifying our product offerings.

We want AI to “just work” for you; we realize how complicated our model and product offerings have gotten.

We hate…

— Sam Altman (@sama) February 12, 2025

Before the launch of GPT-5, OpenAI will release GPT-4.5, known internally as “Orion.” This will be the company’s final model built on traditional GPT technology, which powers the current version of ChatGPT.

The real change will come with GPT-5, which will incorporate all of OpenAI’s technologies, including the capabilities of the unreleased o3 model. This new system will handle everything from creating images to creative writing, complex reasoning, and programming.

Users won’t need to choose between different models for different tasks. Instead, GPT-5 will serve as an AI hub that automatically selects the right approach based on the user’s needs.

The company’s reasoning models currently work through a “chain-of-thought” process, analyzing questions and generating steps before reaching conclusions. These models excel at logical tasks but often struggle with creative work, while creative models like GPT-4.5 face challenges with reasoning tasks.

Growing Competition

In a move that appears to respond to competition from DeepSeek, OpenAI announced that GPT-5 will be available through ChatGPT’s free tier. Users will get unlimited access to the standard intelligence setting, though usage will be monitored to prevent misuse.

The company has created a tiered access system for different user needs. Plus subscribers will have access to a higher intelligence setting with more computing power, while Pro users will get the most capable version of the system.

These changes mean OpenAI will not release its reasoning and coding models separately. Instead, these capabilities will be part of the unified GPT-5 system. The new model will include features such as voice interaction, canvas tools, search capabilities, and deep research functions.

For developers using OpenAI’s API, this change will make integration simpler as they’ll only need to work with one system instead of choosing between different models. However, some specific use cases might be more cost-effective with less computing power and lower per-token prices.

While OpenAI hasn’t provided details about how these changes will affect API costs, Altman expressed confidence in the system’s capabilities. During a recent panel in Berlin, he stated, “I don’t think I’m going to be smarter than GPT-5.”

The move toward a unified system follows the March 2024 development roadmap for GPT-5, which outlined plans for better natural language processing and reasoning capabilities.

When asked about the specific launch timeline on social media, Altman responded with “weeks / months,” suggesting a release in the near future.

weeks / months

— Sam Altman (@sama) February 12, 2025

The post “Single GPT-5 System” Sam Altman Reveals OpenAI’s Plan to Simplify AI Product Line appeared first on Blockonomi.