OpenAI has launched GPT-4.5 as a research preview worldwide. The company claims that this model is more capable of recognizing patterns, making connections, and generating creative insights than previous AI models. Apart from this, GPT-4.5 is also claimed to have a more refined personality, although OpenAI praised this new model, CEO Sam Altman also pointed out some of its shortcomings on X (formerly Twitter).

What is GPT-4.5 and how is it different?

OpenAI said during the live launch that GPT-4.5 has more natural conversations than other models.

Broader knowledge base

Better intent following

Improved emotional intelligence (EQ)

According to OpenAI, GPT-4.5 has been developed in such a way that it reduces the chances of giving "hallucinations".

OpenAI researcher Raphael Gontijo Lopes said, "We designed GPT-4.5 to be a better collaborator, making conversations feel more warm, spontaneous, and emotionally enhanced." According to the company, when GPT-4.5 was compared to GPT-4o, it proved to be better in almost every category.

GPT-4.5 vs OpenAI's 'o' models

OpenAI claims that GPT-4.5 is smarter and more general-purpose than models like OpenAI o1 and o3-mini. The main difference is that GPT-4.5 does not think before giving an answer, which makes it different from reasoning models.

OpenAI believes that in future models, pre-training and reasoning capabilities will work together, although this new model is OpenAI's largest LLM (Large Language Model), it will not be considered a frontier model.

Disclaimer: This content has been sourced and edited from Amar Ujala. While we have made modifications for clarity and presentation, the original content belongs to its respective authors and website. We do not claim ownership of the content.