E-Paper
OpenAI launches GPT-5: What’s new in ChatGPT, availability, and is it free?
GPT-5 is OpenAI's "smartest, fastest and most useful model yet", the AI giant claims. Here's what's new.
OpenAI had been building hype for GPT-5, and today, 7 August, it has finally released the all-new model for everyone. The company is calling it its “smartest, fastest and most useful model yet,” bringing expert-level intelligence to everyone's hands. The good news is that GPT-5 will be available for all users, and for those willing to pay, GPT-5 Pro will be accessible, offering even more advanced reasoning and more detailed, accurate answers.
OpenAI says it has also worked on reducing hallucinations and reducing output that could contribute to biological risk. In this regard, OpenAI says it has implemented strong safeguards to minimise risks.
Here is everything you need to know about the GPT-5 family, its availability and more
GPT-5 Availability And Is It Free To Use?
Before we dive into the technical details, let's discuss availability. GPT-5 will be available for all Plus, Pro, Team and Free users. This means that if you are using the free version of ChatGPT, you will be able to access GPT-5. The company says that access for Enterprise and Edu customers will be available in a week's time.
Regarding usage limits, paid users can use GPT-5 without restriction and will also get access to GPT-5 Pro. Team, Enterprise and Edu customers can also use GPT-5 as the default model for their everyday tasks.
For free users, the company states there is a limit: once they reach their GPT-5 usage limits, ChatGPT will revert to GPT-5 Mini, which is a smaller but still quite capable model.
As for availability, the rollout of GPT-5 starts today, 7 August. However, the company says that for free users, it could take a few days for the full reasoning capabilities to become available.
How To Access GPT-5?
You do not have to do anything as whenever you open ChatGPT while being signed in, the model you use will be GPT-5. It will replace GPT-4o, OpenAI o3, OpenAI o4-Mini, GPT-4.1 and GPT-4.5. All you need to do is type a prompt, and GPT-5 will take over from there.
GPT -5: What Is It Capable Of?
The company says that GPT-5 is a significant leap in intelligence compared to its previous models, across all areas like coding, maths, writing, health and visual perception. It also has the ability to understand when a quick response is needed and when it should take more time to provide a detailed answer.
OpenAI also says that GPT-5 outperforms all previous models in benchmarks and provides answers more quickly, while at the same time being more useful and reliable for real-world queries. The company has also worked on reducing hallucinations, improving instruction following and boosting ChatGPT's performance in three key areas: writing, coding and health.
When it comes to coding, OpenAI claims this is, as expected, its strongest model to date. The company says that you can create small games, such as a rolling mini-ball game, a typing game, a drum simulator and a lo-fi visualiser, with single prompts.
ChatGPT will now also be better at writing and editing-based tasks. It will now be better at tasks such as editing reports and working with emails. Furthermore, there will be a significant focus on providing better answers to health queries. The company says that it scores higher than any previous model in benchmarks, but it is not designed to replace a registered medical practitioner. Instead, it is portrayed as a partner that helps you to understand your health better.
Comparison To Older Models Like GPT-4o
Apart from being less susceptible to hallucination, the company says that GPT-5's responses are 45% less likely to feature a factual error compared to GPT-4o. When reasoning, GPT-5's responses are 80% less likely to contain a factual error compared to OpenAI's O3. Meanwhile, GPT-5 Pro replaces OpenAI's O3-Pro and, as expected, performs the best in the GPT-5 family across various benchmarks.
Biological harm is another area to which many LLMs have been accused of contributing. This is why GPT-5's reasoning has a built-in safety mechanism with a multi-layered defence system for biology.
MOBILE FINDER: iPhone 16 LATEST Price
