Model Compare

OpenAI has introduced GPT-6.1 Terra, publishing the announcement on its official site. The news is spreading quickly among developers and AI observers, with discussion focused on what the new model offers and how it compares to previous releases. Specific details about capabilities or availability have not yet been widely reported, and reactions are still forming as people read the announcement.

Samsung's Galaxy Z Fold 8 is being discussed as anticipation builds around the company's next foldable smartphone. The buzz comes alongside coverage of competing foldables, including reports that some early buyers of Apple's new foldable iPhone said the device broke within hours of purchase. Readers are weighing Samsung's foldable lineup against Apple's entry into the category.

Google DeepMind has announced Genie 4, described as a new generation of world simulation, in a post on its research blog. The announcement is drawing heavy attention among AI researchers and enthusiasts, who are debating what the model can do, how it compares with earlier generative world models, and what it means for gaming, robotics and simulation research.

A large-scale study drawing on 500,000 electric vehicle battery health checks has compared 20 EV models, with one car emerging as the clear standout for battery condition. Battery degradation is a top concern for used EV buyers, so findings on which models hold their health best are drawing attention from drivers and the auto industry.

Google's Gemini 4 Carbon has taken the number one position on the Artificial Analysis Intelligence Index version 5, becoming the highest-ranked model on the independent AI benchmark. The result prompts discussion among AI watchers about how Google's newest model compares to rivals like OpenAI and Anthropic on measured intelligence, reasoning quality, and whether independent leaderboards can keep pace with rapid model releases.

The Verge has published a look at the keyboards used by its own staff, showcasing the setups, models and custom builds favoured by the outlet's writers and editors. Keyboard culture among tech journalists attracts attention because mechanical keyboards have become a serious hobby, with enthusiasts comparing switches, keycaps and layouts as much as the devices they type on.

Norway is moving ahead with plans to expand and modernise its duty-free retail offer, positioning its travel shopping model as a benchmark that could outdo comparable offerings in the United States. The travel retail industry is watching closely, with commentators noting the ambition behind the Nordic country's plans to make its airport shopping more attractive to passengers.

Artificial Analysis has published version 5 of its Intelligence Index, a benchmark that compares leading AI models on reasoning, knowledge and problem-solving performance. The update ranks the latest frontier models from major labs and is drawing attention as developers and researchers look for independent guidance on which systems currently lead the field.

GPT-6.1 Terra (Max), an OpenAI model, has recorded a score of 48 on Artificial Analysis, a benchmarking site that compares AI models on capability. The result was shared in a single-sentence note on Hacker News, drawing modest attention. No further details about the benchmark methodology, comparison with rival models, or OpenAI's response were included.

A new benchmark from Native claims that storing Markdown content in SQLite can cut token consumption for AI workflows by roughly 40 times compared with raw Markdown files. The post outlines a method for serving structured content to language models with far less context overhead. Readers on Hacker News are debating the claim, questioning how token savings were measured and whether the gains hold up outside narrow benchmarks.

Astronomers say most planetary systems discovered beyond our own differ sharply from the solar system, with planets arranged in configurations not seen around the Sun. An explainer by BBC Science Focus Magazine outlines why our arrangement of rocky inner planets and gas giants appears to be the exception, citing patterns found in exoplanet surveys and models of how planets form and migrate.

A new iPhone model referred to as the iPhone Duo is being compared against other iPhone sizes, with questions raised about whether it will fit comfortably in a pocket. The discussion centers on the device's dimensions relative to existing iPhones, as buyers weigh portability against screen size. Exact specifications have not been confirmed, and reactions so far focus on curiosity about the form factor rather than confirmed details.

Tom's Guide has published a hands-on comparison of Apple's iPhone Duo and iPhone 18 Pro Max, testing both devices and concluding with a recommendation on which one buyers should choose. The piece compares performance, features and value across the two models. As one of the first detailed reviews of the new lineup, it is drawing attention from consumers weighing an upgrade ahead of purchase decisions.

A lifetime deal costing $30 lets users compare answers from multiple AI chatbots — including ChatGPT and Claude — without paying separate monthly subscriptions. The offer targets people frustrated by stacking recurring fees for individual AI services, promising a single one-time payment for access to several models side by side.

Stellantis performance division SRT says it has solved the turbo lag problem affecting the twin-turbo Hurricane engine, using a new technology detailed by Motor1.com. The Hurricane six-cylinder engine, used across Dodge, Jeep and Ram models, has faced criticism over throttle response compared with the V8s it replaces. SRT's announcement suggests the performance division sees the engine as viable for future high-output vehicles.