Join Forward-thinking Leaders
Elevate your expertise with tech insights, startup breakthroughs, and leadership intelligence curated for your priorities.
OpenAI Unveils Most Capable o3-Pro Model: Check Details
OpenAI- an AI tool, widely used by professionals to make their workflow easier and management smoother. It is not only saving time by completing recurring tasks but also helping people in making the various processes automated to reduce manual labor. But it keeps updating itself to provide the best services and experience.
Summary:
1. OpenAI has unveiled its latest AI model, o3-pro, aiming to set a new standard for AI analysis.
2. This latest model is expected to replace o1-Pro for ChatGPT Pro and Team users, followed by Enterprise and Edu users.
3. The model includes web search, Python code execution, file analysis, image interpretation, and memory-based personalized answers.
So, following this, it has recently unveiled its most capable AI model. The name speaks louder about this AI model. Hence, we will walk you through its features in detail and will discover how it stands out among other AI tools.
The o3-pro model belongs to OpenAI’s novel category of “reasoning models”- AI systems developed to solve issues step by step, making them more dependable in technical domains like physics, programming, and complex mathematics.
It has officially introduced its most powerful model so far, o3-pro, to set a new standard for AI analysis. From this week, the new version will replace o1-pro for ChatGPT Pro and Team users, followed by Enterprise and Edu users. Developers may also get it right away using the API, which costs $20 per million input tokens and $80 per million output tokens- roughly 750,000 words for every million input tokens.
What is o3-Pro Model?
The o3-pro model is one of OpenAI’s latest “reasoning models”- AI systems that solve issues step by step, making them more dependable in technical fields such as physics, programming, and sophisticated mathematics. Internal reviews reveal that it outperforms the old version, o3, in all evaluated categories, particularly science, business, education, and writing support.
OpenAI reports that testers complimented its clarity, accuracy, and ability to follow instructions.
“This o3-pro model has access to tools that make ChatGPT useful- it can search the web, analyze files, explain about visual inputs, use Python, personalize responses using memory, and more,” claimed OpenAI in an X.org article.
Sam Altman, CEO of OpenAI, said the model is quite intelligent. “I didn’t believe the win rates relative to o3 the first time I saw them,” Altman wrote on X.
In terms of features, o3-pro includes web search, Python code execution, file analysis, image interpretation, and memory-based personalized answers. Unfortunately, it is currently unable to create graphics, does not support OpenAI’s Canvas workspace feature, and, because of a technical glitch, ChatGPT does not support temporary conversation.
Another disadvantage: answers are substantially slower than o1-pro.
In terms of performance, the o3-pro appears to be the clear leader. It allegedly outscored Google’s Gemini 2.5 Pro in AIME 2024 (a math benchmark) and Anthropic’s Claude 4 Opus in GPQA Diamond, a test designed to assess PhD-level science understanding, according to OpenAI.
Similarly, Meta is making headlines to make a massive investment of $10 billion in scale AI. The goal of this investment is to introduce modern training features and to make it efficient and effective. These are those startups that lead the AI data market.
The startup was valued at $14 billion earlier in 2024 as part of a funding round including investors such as Meta. Scale’s evolution has both shaped and been shaped by the AI boom that OpenAI unleashed. Scale’s early focus was on labelling photos of autos, traffic lights, and street signs to help train self-driving car models.
Follow The Techrising to stay ahead with the latest stories in startups, tech, so you never miss a breaking story or trending headline.
