OpenAI rolls out new GPT-5.2 AI model to counter Google’s Gemini push

Key Takeaways

• OpenAI has launched GPT 5.2 with enhanced features for developers and professionals.
• GPT 5.2 offers three specialized versions and sets new benchmarks in coding, reasoning, and real-world tool use.

OpenAI released GPT 5.2 on Thursday, calling it its most advanced model to date and positioning it for both developers and everyday professional use as pressure from Google increases.

The launch comes as competition heats up with Google’s Gemini 3, which has set new benchmarks in reasoning and multimodal performance.

OpenAI reportedly fast-tracked GPT-5.2’s release in response, aiming to match or exceed Gemini’s capabilities in key areas like long-context analysis, coding, and tool use. Early evaluations suggest GPT-5.2 may hold an edge in complex reasoning, as both models push for dominance in enterprise and developer AI workflows.

GPT 5.2 will be available to ChatGPT paid users and through the API in three versions: Instant for fast everyday tasks such as writing, information requests and translation, Thinking for structured work such as coding, long document analysis, math, and planning, and Pro for maximum accuracy and reliability on difficult problems.

“We designed 5.2 to unlock even more economic value for people,” said OpenAI chief product officer Fidji Simo, noting improvements in creating spreadsheets, building presentations, writing code, image processing, and coordinating multi-step projects.

OpenAI says GPT 5.2 delivers new benchmark records in coding, math, science, vision, long context reasoning, and tool use, enabling more reliable agent workflows, production-grade code, and complex systems operating across real-world data.

Source