"The new model, available in Instant, Thinking, and Pro performance tiers, offers major improvements across a range of benchmarks, the company said. Using OpenAI's GDPval benchmark, which compares the model's ability to complete 44 different business tasks to the same standards as human experts, GPT-5.2 matched or exceeded human users in 70.9% of tests, compared to GPT-5.1's 38.8% across the Instant (basic), Thinking (deeper reasoning), and Pro (research-grade) versions."
"To illustrate these advances, OpenAI said that GPT-5.2 Thinking could fully format a workforce planning spreadsheet, while on GPT-5.1, the equivalent output assembled the same spreadsheet correctly, but in a more basic state that lacked formatting. "We designed GPT‑5.2 to unlock even more economic value for people; it's better at creating spreadsheets, building presentations, writing code, perceiving images, understanding long contexts, using tools, and handling complex, multi-step projects," said OpenAI."
OpenAI released GPT-5.2 with notable performance gains across Instant, Thinking, and Pro tiers. GPT-5.2 matched or exceeded human users on 70.9% of GDPval business-task tests versus GPT-5.1's 38.8%. The model shows tangible improvements in spreadsheet formatting, presentation creation, code writing and debugging, image perception, long-context understanding, tool usage, and handling complex multi-step projects. GPT-5.2 also recorded gains on benchmarks including ARC-AGI-1/ARC-AGI-2 and SWE-Bench Pro/SWE-Bench Verified, enabling more reliable production debugging, feature implementation, refactoring, and end-to-end fixes with less manual intervention. Rollout began to ChatGPT paid plans and API access with specified token pricing and discounts.
Read at InfoWorld
Unable to calculate read time
Collection
[
|
...
]