OpenAI has responded to Google’s Gemini 3 Professional transfer with GPT-5.2. Listed here are the improvements supplied by GPT-5.2, that includes superior accuracy, stronger evaluation capabilities, and three new mannequin choices…
On the identical day it introduced the Sora licensing settlement with Disney, OpenAI launched its new mannequin, GPT-5.2. The corporate describes this mannequin as essentially the most competent system for skilled use in actual life to this point. OpenAI acknowledged that the mannequin affords vital progress in areas comparable to “getting ready spreadsheets, creating displays, writing code, visible notion, understanding lengthy content material, utilizing instruments, and executing multi-step initiatives.”
Superior Efficiency and Benchmark Outcomes

In response to 10 completely different benchmark exams shared by OpenAI, essentially the most superior model, GPT-5.2 Pondering, surpassed GPT-5.1 by a large margin in lots of areas.
Within the AIME 2025 take a look at, it achieved 100% success by appropriately answering all 30 difficult math questions. This consequence exceeded GPT-5.1’s 94% efficiency and was achieved with out utilizing instruments.Within the ARC-AGI-1 take a look at, a rise of greater than 10 factors was noticed in summary considering capability in comparison with GPT-5.1.
Extra Correct Responses and Suitability for Skilled Use
OpenAI states that GPT-5.2 Pondering makes 30% fewer errors when answering factual data. This improvement makes the mannequin extra dependable in analysis, writing, evaluation, and choice help processes.
Enhanced Dialog Expertise

GPT-5.2 Prompt, the model customers will encounter most continuously, has been optimized for pace and practicality. In response to OpenAI, the mannequin has been considerably improved in areas comparable to:
Info looking out,How-to content material,Technical explanations,Guides,Translation and textual content era.
Moreover, the hotter and extra fluid conversational tone that got here with GPT-5.1 Prompt has been preserved and improved.
Expectation and Aggressive Strain Following GPT-5
Launched in 2025, GPT-5 didn’t meet person expectations and acquired criticism for weak responses and a “boring character.” This dissatisfaction even led to calls for for the return of GPT-4o.
Throughout this era, Google’s Gemini 3 Professional mannequin settled on the high of the LMArena rating, whereas OpenAI’s GPT-5.1 mannequin was in sixth place. Following this case, it was reported that Sam Altman’s staff was directed to a brand new enchancment course of with “code purple” precedence.
Expectations with GPT-5.2

OpenAI goals to distinguish itself from opponents in scale with agreements signed for over $1.4 trillion in infrastructure investments. In a notice despatched to workers, Altman acknowledged that GPT-5.2 can be on the identical stage as Gemini 3 Professional. With the mannequin’s launch, how this declare will play out stays a matter of curiosity.
Three Completely different Variations Being Rolled Out
OpenAI is providing GPT-5.2 with three choices:
GPT-5.2 InstantGPT-5.2 ThinkingGPT-5.2 Professional
These variations will first be opened to paid plan customers. Moreover, GPT-5.1 will stay accessible within the “older fashions” part for 3 months.
You May Additionally Like;
Observe us on TWITTER (X) and be immediately knowledgeable concerning the newest developments…

