Robotics

Canon EOS R100 Ant Pruitt/ZDNETZDNET’s key takeawaysThe Canon EOS R100 More

Matthew Miller/ZDNET ZDNET’s key takeaways The H2O Audio Tri 2 More

The iPhone 16 Pro Max and 16 Pro in the Steve Jobs Theater on the Apple Park campus. Kerry Wan/ZDNETI spent several hours with Apple’s new flagship iPhone 16 Pro More

Adrian Kingsley-Hughes/ZDNET ZDNET’s key takeaways The Blackview Mega 1 is available directly from Blackview More

Sabrina Ortiz/ZDNETIf you have used ChatGPT, you know that the chatbot outputs answers incredibly quickly, taking seconds to process even complex queries. Although speed is a clear advantage, it can also mean the chatbot rushed through generating an answer. These new OpenAI models specialize in tackling that issue. Also: Gemini Live is rolling out to all Android users – for free. How to access itOpenAI unveiled OpenAI o1 on Thursday, a new series of models designed to work through more complex science, coding, and math problems by spending more time thinking before they respond, according to the blog post. We’re releasing a preview of OpenAI o1—a new series of AI models designed to spend more time thinking before they respond.These models can reason through complex tasks and solve harder problems than previous models in science, coding, and math. https://t.co/peKzzKX1bu— OpenAI (@OpenAI) September 12, 2024

OpenAI shares that it trained the models to think before responding, like humans do, refining their thinking process and allowing them to try different strategies and identify their mistakes. This approach has paid off, with the o1 model excelling in math and coding, scoring 83% on the International Mathematics Olympiad (IMO) qualifying exam. For comparison, GPT-4o correctly solved only 13% of problems. Open AI CEO Sam Altman highlighted some of the benchmark results in an X post, seen below.  The results make sense, given that a popular way to make ChatGPT output higher-quality responses, especially with prompts requiring advanced reasoning, is requesting it to reread the prompt. When reprocessing the original request, it typically finds its error and outputs the correct response. Also: How ChatGPT scanned 170k lines of code in seconds and saved me hours of workBecause o1 is an early model, it lacks key ChatGPT features, such as internet browsing and accepting media uploads. As a result, in the short term, GPT-4o may be the best model for common cases, while o1 will be a better option for solving complex science, coding, and math problems. OpenAI also launched o1-mini, which is 80% cheaper than o1-preview. This makes it a more cost-effective and faster alternative for developers. OpenAI shares in the blog post that o1-mini is specifically effective at coding.  More

Kerry Wan/ZDNETiPhone 16 preorders opened up at 5 am this morning, and by the time I had enough coffee to remember my name, the planned September 20 delivery date had already been exhausted. More

Get three months of access to Xbox Game Pass Ultimate for a discount right now. Stack Social If you’ve been thinking about trying out Xbox’s subscription gaming service Game Pass, now is a great time to get started: Stack Social is offering a deal to get three months of Xbox Game Pass Ultimate access for […] More

It looks like a standard laptop, but it’s not. Jack Wallen/ZDNETZDNET’s key takeawaysThe CrowView Note More