Topic: ai model
-
Google Tests Its Fastest AI Image Model Yet
Google is developing a new, high-speed image generation model called "Nano Banana 2 Flash", designed for fast, affordable, but less complex tasks compared to its more powerful sibling. The flagship model, "Nano Banana Pro (Gemini 3 Pro Image)", is built for demanding creative projects, offeri...
Read More » -
Google's Gemini 3 Launches, Already Fuels AI Mode
Google has launched Gemini 3, its most advanced AI model, integrating all Gemini abilities into a single tool and making it available in Google Search from day one. The model enhances Search with complex reasoning and real-time generative features, such as producing interactive visualizations and...
Read More » -
AI Photo to 3D World: Explore Your Images Like Never Before
Tencent's HunyuanWorld-Voyager is an AI model that generates immersive, navigable 3D-like video sequences from a single still image, eliminating the need for traditional 3D modeling. The system produces 2D video frames with depth information to simulate camera movement, creating up to two-second ...
Read More » -
UAE Unveils Compact Yet Potent AI Model
The UAE has launched K2 Think, a sovereign open-source AI model that rivals leading U.S. and Chinese systems in reasoning capabilities despite using fewer parameters. Developed by Mohamed bin Zayed University, K2 Think specializes in complex problem-solving through simulated deliberation and is o...
Read More » -
AI Model Identifies Plants with Precision
Carbon Robotics has developed a Large Plant Model (LPM), an AI trained on over 150 million images that can instantly identify plant species, eliminating the need for lengthy manual retraining of its robotic weeders. The system allows farmers to instruct robots in real-time to target new weeds wit...
Read More » -
Google's Gemini 3 Makes Search Smarter
Gemini 3 is Google's advanced AI model that enhances reasoning, multimedia interpretation, and coding within its ecosystem, starting with Search to make digital interactions more intuitive and efficient. Google is embedding AI across its product lineup, from established services like Maps and Gma...
Read More » -
Canva’s ‘Creative OS’ Is a Marketing Workspace
Canva is expanding its platform with a suite of new digital marketing and video-editing tools, centered around a proprietary AI model and branded as a "Creative Operating System" for comprehensive team workflows. The update includes a redesigned video editor for accessibility, a new Forms tool fo...
Read More » -
ServiceNow and Nvidia Launch Open-Source Security Model
ServiceNow and Nvidia have launched Apriel 2.0, an open-source AI model designed for building custom agents and enhancing security operations in regulated environments. Built on Nvidia’s Nemotron architecture, it integrates with ServiceNow’s compliance-certified platform, allowing seamless use of...
Read More » -
DeepSeek's New AI Model Challenges Alibaba Qwen and OpenAI
DeepSeek has launched an experimental AI model, DeepSeek-V3.2-Exp, challenging competitors like Alibaba and OpenAI with its advanced Sparse Attention technology that cuts computational costs and boosts performance. The company introduced a 50% reduction in API pricing to lower adoption barriers a...
Read More » -
Physical Intelligence's new robot brain learns untaught tasks
A San Francisco robotics startup has developed an AI model, π0.7, that demonstrates "compositional generalization", enabling robots to combine learned skills to tackle novel tasks they were never specifically trained for, such as operating an unfamiliar air fryer. The model shows a promising ...
Read More » -
Step Into AI Videos: OpenAI's Sora 2 Adds Sound and You
OpenAI has launched Sora 2, a video-generation model that integrates synchronized audio and visual effects for the first time, alongside a dedicated iOS app for user cameos. The model demonstrates significant improvements in visual consistency, handling complex multi-shot instructions and simulat...
Read More » -
One AI Model Unlocks Humanlike Robot Movement
The Atlas humanoid robot now uses a single, unified AI model for both movement and manipulation, enabling more fluid, humanlike behavior and unexpected emergent skills. Developed by Boston Dynamics and Toyota Research Institute, this large behavior model processes visual, proprioceptive, and lang...
Read More » -
Google Drive adds ransomware detection and file recovery
Google has launched ransomware detection and file restoration for Google Drive, moving from beta to general availability to help businesses mitigate malware threats. The system uses AI-powered scanning via Google Drive for desktop to proactively identify ransomware, alerting users and administrat...
Read More » -
Nvidia DLSS 4.5 Boosts Frame Generation 6x, Enhances Image Quality
Nvidia has released DLSS 4.5, a major upgrade to its AI upscaling technology that improves image quality and performance for all GeForce RTX graphics cards. The update features a new, more powerful AI model that significantly reduces visual artifacts and a new 6x Multi Frame Generation mode, but ...
Read More » -
Anthropic Acquires Bun as Claude Code Hits $1B Milestone
Claude Code, an AI-powered software development platform, has achieved a $1 billion revenue run-rate and is being enhanced by Anthropic's acquisition of the high-performance JavaScript runtime, Bun. The acquisition will directly improve Claude Code with faster performance, greater stability, and ...
Read More » -
Apple Strikes $1 Billion AI Deal With Google to Power Siri
Apple is negotiating a billion-dollar annual deal to license Google's 1.2 trillion-parameter AI model to upgrade Siri's capabilities. The partnership aims to enhance Siri's performance by integrating Google's advanced AI infrastructure, marking a significant shift in AI sector dynamics. Both comp...
Read More » -
Apple execs to discuss China, tariffs but not foldable iPhone: Kalshi traders
Apple's upcoming earnings call, following Tim Cook's announcement to step down as CEO in September, will be closely watched by Kalshi traders who are betting on key discussion topics. Traders assign near-certain odds (98% and 96%) that Apple will discuss China and tariffs, while other potential t...
Read More » -
Ex-Google X Trio Raises $6M to Build Your AI Second Brain
TwinMind is an AI-powered app that captures ambient speech to build a personal knowledge base, operates entirely offline, and recently secured $5.7 million in seed funding. The app distinguishes itself by running unobtrusively in the background on mobile devices, offering real-time translation an...
Read More » -
Google Meet and Translate get Gemini 3.5 Live Translate with listening mode
Google has launched Gemini 3.5 Live Translate, featuring a new listening mode that enables real-time, speech-to-speech translation in Google Meet and Google Translate without manual toggling. The update provides live captions and audio translations in Google Meet for smoother multilingual meeting...
Read More » -
ChatGPT Takes on Google Translate: AI Translation Battle
OpenAI has launched a dedicated web-based translation service called ChatGPT Translate, directly challenging Google Translate and supporting over fifty languages with a clean interface. Unlike Google Translate, which handles images, documents, and web pages, ChatGPT Translate currently focuses on...
Read More »