GPT-5.2 raises accuracy and speed, with 256K token context support, so you get clearer answers on long files and chats.
After 10 text and 4 image tests, OpenAI's latest model barely beats GPT-5.1. What are Plus subscribers really getting?
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback