GPT-5.2 launch. A new series of models built for specialized tasks
OpenAI is rolling out the GPT-5.2 series, built for professional work. According to OpenAI, the new models handle coding, data analysis and long documents better, while making far fewer mistakes.
A month after launching GPT-5.1, OpenAI is rolling out the GPT-5.2 series (Instant, Thinking, and Pro), which it calls its most advanced solution for professional work. The tool’s creators emphasize it’s moving beyond simple chat to tackle complex, multi-step tasks. The new models are designed to deliver real economic value—from building advanced spreadsheets and writing code to precise analysis of images and long texts.
Expert-level performance
In the GDPval tests, which cover tasks typical of 44 different occupations (representing key sectors of the U.S. economy), the GPT-5.2 Thinking variant matched or outperformed human experts in 70.9% of cases. For comparison, the previous version (GPT-5) hit that mark in only 38.8% of trials.
Share
You might be interested in
-
Mistral 3: unveiling the new generation of open-weight models
Mistral AI is introducing the Mistral 3 model family, which includes the Large version with MoE architecture and the Ministral series for local devices. All models are available under the Apache 2.0…
-
Claude Opus 4.5: a new standard in coding?
Anthropic has released Opus 4.5, a model designed to outperform engineers in technical tests, all while costing just a fraction of its predecessor’s price.
-
Nano Banana Pro: Precision, text and verification
Google has launched Nano Banana Pro — a model based on the Gemini 3 Pro architecture. This new gadget brings advanced text handling, lighting control and built-in SynthID verification.
-
Is Grok getting softer? New model from xAI
xAI has rolled out Grok 4.1, their latest language model update. The company boasts significant improvements in understanding intentions, speech coherence and error reduction, all aimed at enhancing interactions with users.
