OpenAI has officially released GPT-5, which it is marketing as its “smartest, fastest, and most useful model yet” with integrated reasoning capabilities that it claims dwarves rival models. The new model is available to all users, including free tier, through ChatGPT and OpenAI’s API, with variants like GPT-5, GPT-5-mini, GPT-5-nano, and GPT-5-chat designed for different use cases. How does GPT-5 compare to its contemporaries? News of GPT-5’s launch was announced on Thursday, and it is expected to excel especially across domains like writing, coding, and health care. According to OpenAI, GPT-5 has lower hallucination rates, meaning the model fabricates answers less frequently. To make sure no mistakes were made, the company said it carried out extensive safety evaluations during the model’s development, and this included 5,000 hours of testing. OpenAI has said GPT-5 will not outrightly refuse to answer questions that are potentially risky; instead, it will use “safe completions,” which means it will provide high-level responses within safety constraints that can’t be used to cause harm. During the briefing, OpenAI also demonstrated GPT-5’s “vibe coding” abilities, asking the model to create an engaging web app that could help an English speaker learn French. Aside from its engaging theme, the app also had to include activities like flashcards and quizzes, as well as a way to track daily progress. The new model was tested with the same prompt entered into two different windows, and it generated two different apps within seconds. The apps were not perfect, according to an OpenAI lead. However, users will be able to make additional tweaks to the AI-generated software, including toggling the background or adding additional tabs. Box, a company that manages computer files for companies, has been vigorously testing GPT-5 across a wide variety of data sets in recent weeks, and its CEO, Aaron Levie, said it succeeded where many previous AI models have failed, even going as far as to tag it a “complete breakthrough.” Despite all this, there are those who believe GPT-5 is still less superior to Grok 4, touted to be the most powerful AI on earth. Naturally, Musk is one of them, but he has the backing of folks like François Chollete, creator of ARC-AGI, who has highlighted Grok 4 beating GPT-5 on the ARC-AGI interface. Comparison of AI models. Source: @fchollet (X/Twitter) Musk also said on X that a newer Grok model , Grok 5, would be released before the year’s end, and he promises it will be even better. Bottom line though: Grok 4 Heavy was smarter 2 weeks ago than GPT5 is now and G4H is already a lot better. Let that sink in. https://t.co/BrggsEwnuz — Elon Musk (@elonmusk) August 7, 2025 Microsoft has incorporated GPT-5 into its products According to reports, Microsoft officially incorporated OpenAI’s GPT-5 into its consumer, developer, and enterprise products on Thursday, August 7, 2025. The integration of GPT-5 is expected to enhance various Microsoft services, including Microsoft 365, Azure, and Teams, by providing advanced AI capabilities for users. Users subscribed to Microsoft 365 Copilot and Microsoft Copilot, enterprise and consumer packages, can automatically enjoy GPT-5. So will developers across Microsoft platforms. The company has tested the new model, and the results show that GPT-5 has one of the strongest AI safety profiles among previous models against several modes of attack, including malware generation, fraud/scam automation, and other harms. Want your project in front of crypto’s top minds? Feature it in our next industry report, where data meets impact.
