OpenAI has unveiled GPT-4.5, its latest and most advanced language model to date. This release brings several significant enhancements over its predecessors, aiming to provide users with a more natural and efficient AI experience.
The latest model features upgraded chat capabilities, greater computational efficiency, and more. Here are the five most notable upgrades in ChatGPT-4.5 that are currently available to ChatGPT Pro users and rolling out next week to ChatGPT Plus and Team users.
1. More natural conversations
OpenAI is constantly enhancing its chatbots' ability to communicate with its users naturally, and GPT-4.5 has been fine-tuned to deliver more human-like interactions. Thanks to added emotion, users will notice that conversations feel more intuitive and engaging. This model demonstrates a refined personality and an improved understanding of context. The advancement makes it particularly adept at tasks requiring a nuanced grasp of language, such as creative writing and problem-solving.
2. Expanded world knowledge
Building upon its predecessors, GPT-4.5 offers a broader and deeper understanding of the world. This expanded knowledge base allows it to provide more accurate and relevant information across various topics, enhancing its utility for users seeking detailed insights into content from news stories to history lessons.
3. Reduced hallucination rate
One of the critical improvements in GPT-4.5 is its decreased tendency to generate inaccurate information, commonly referred to as "hallucinations." Early testing indicates a significant reduction in hallucination rates compared to previous models, making GPT-4.5 a more reliable tool for users who require precise and factual responses.
4. Optimized performance
Despite being OpenAI's largest model to date, GPT-4.5 has achieved over a tenfold increase in computational efficiency compared to GPT-4. This enhancement means users can expect faster response times and a more seamless experience, even when handling complex queries.
5. Advanced multimodal capabilities
GPT-4.5 extends its functionality beyond text by accepting image inputs and generating inline SVG images. It also integrates with DALL-E for image generation, allowing users to create visual content directly through the model. Additionally, GPT-4.5 supports web search, ensuring access to the most up-to-date information available when users utilize the search feature.
Final thoughts
These upgrades position GPT-4.5 as a versatile and powerful chatbot that can take on various tasks, from creative storytelling and image generation to detailed research, summaries, and problem-solving. While the chatbot is currently only accessible to users at the Pro level, Plus users can anticipate seeing the model in their drop-down options as ChatGPT-4.5 rolls out in the next week or so.
More from Tom's Guide
- Alexa+ — here's everything you need to know
- I used Claude's new AI model to design my dream home — here's what happened
- What is 'Gibberlink'? Why it's freaking out the internet after these two AIs talking to each other went viral
You must confirm your public display name before commenting
Please logout and then login again, you will then be prompted to enter your display name.
