According to the Vectara platform, which is the grading of error frequency models, the new AI is really less wrong (1.4%against 1.4%), but just a while in the GPT-4O (1.49%).
Interestingly, the ChatGPT-4.5 preview showed an even lower indicator-1.2, and 0.795%O3-Mini remained the leader of the leader.
Despite improvements, the GPT-5 is still lower than 4.5 accuracy, but although the Gemini-2.5-pro (2.6%) and Grok-4 (4.8%) (4.8%) are significantly ahead of competitors.
The launch caused a scandal because Openai lifted the old GPT-4, 4 and 4O-Mini models without warning from plus subscriptions. Many users have complained that they have lost their usual functions, and some call it “loss of one friend”.
Sam Altman accepted the importance of old versions for the audience and promised to temporarily return the GPT-4O for subscribers to evaluate the “inheritance” request.
Source: Ferra
I am a professional journalist and content creator with extensive experience writing for news websites. I currently work as an author at Gadget Onus, where I specialize in covering hot news topics. My written pieces have been published on some of the biggest media outlets around the world, including The Guardian and BBC News.
