Google

On December 6, according to news, yesterday, Google officially launched a new generation of reasoning models, Gemini 3 Deep Think, known as the “IMO gold-class” model, with a significant leap forward from the previous generation, Gemini 2.5 Deep Think, leading in many international benchmarking tests. Gemini 3 Deep Think uses a "parallel reasoning" structure, which simultaneously explores a variety of assumptions, thus demonstrating a breakthrough in complex mathematical, scientific and logical dilemmas. In Hum..

On December 5th, according to TechCrunch, Google is testing an important update on a global scale to integrate the search 's AI overview with the AI mode, with the aim of allowing users to " seamless access " to the AI dialogue mode on the search result page. This test is first open to mobile users. After seeing the AI overview in the search results, the user can enter the AI mode directly through the new "Ask AI" button and continue with multiple rounds of dialogue without having to enter the AI 2 interface separately. Google Search Product..

On November 6th, according to the media TestingCatalog, a new up-to-date alert card appeared in the previous Google Gemini web interface. According to its analysis, Google is preparing a new image generation model for the nano Banana series. It was described that the model would be named "GEMPIX2" and published as a second generation of nano Banana. The report analysis, the advance card, is usually a proximate signal, and, according to previous assumptions, such updates usually appear in public..

On November 6th, Google officially launched the data centre landing program, with the intention of putting all its calculations in space. It's called Project Suncatcher, Solar Catcher Project. Google's idea is simple: instead of fighting for depleted resources on Earth, it's better to go to space with solar energy. This whole new moon landing programme has only one goal, creating a solar-powered, scalable AI infrastructure in space. Specifically, Google, in order to solve the energy problems of the data centres, their plans..

On October 16th, Google officially released Veo 3.1 Video Generation Model Updates today, with significant improvements at both functional and model levels. In terms of functionality, for the first time, Veo 3.1 added audio support to the "Reflection Video" and "Suspend" functions to make the creation process more complete. Users can not only define roles and styles through multiple reference images, but can also use first-end images to generate seamless transitions, or extend to produce content in excess of one minute. At the model level, Veo 3.1 has made significant progress in the understanding and audio-visual quality of the hints..

Early this morning, Google officially presented its most advanced image generation and editing model - Gemini 2.5 Flash Image (aka nano banana). According to the official introduction, the main features of Gemini 2.5 Flash Image include the following: Fully maintain the consistency of the character: it can easily place the same character in different environments, or show the same product from multiple angles, while perfectly keeping its core subject unchanged. Cue-based image editing: Allows the user to use simple natural language...

On Monday, Google announced that its DeepMind had won the AI Olympiad Gold Medal (IMO) and received official certification, a crazy slap in the face of OpenAI, and yesterday, according to CNBC and TBPN, the DeepMind team has been poached by a number of members in quick succession. CNBC reported that Microsoft has recruited about 20 employees from Google's DeepMind team in recent months. Among them, Amar...

In the past, our impression of AI-generated video was, at best, an awkward collage of "motion picture level", with no sound, no atmosphere, no emotion, like a puppet show. Now, Google has released a new video generation model - Veo 3 - that directly shatters this standard and rebuilds it. Now what you see is no longer a cold clip, but a real "talking" "emotional" anthropomorphic video generation. In this post, I'm going to take you on a deep dive into how Veo 3 works. At the same time, at the end of the article, I will also teach you: how to use AI to write video to generate prompts, to help you improve...

Recently, Google Gemini 2.5 Pro announced the release of a new version "Gemini 2.5 Pro(0605)". In terms of performance, the new version of Gemini 2.5 Pro has stronger performance in code, reasoning and other tasks: in Humanity's Last Exam, it outperformed OpenAI o3 with 21.6%; in GPQA test, it topped the list with 86.4%. In the Big Model Arena, Gemini 2.5 Pr...

A few days ago, Google co-founder Sergey Brin was interviewed by the company, in which he mentioned quite a lot about the present and future of AI. When discussing the topic of "current AI development versus past expectations," Brin said that the sudden progress of AI from "playful imagination" to today's display is surprising. He said that back then, people's predictions of the future were like a theoretical 'chat' that no one could predict, but today's AI breakthroughs have made everything 'surprisingly real'. With the likes of the Gemini 2....

Google NotebookLM is an AI-powered knowledge management app that helps users transform complex information into clear and understandable content. Whether you're a student, researcher or professional, you can use NotebookLM to enhance your learning and productivity, and learn and understand information in a variety of ways. By uploading a variety of materials and asking questions, users can quickly access key information and learn through AI-generated audio overviews. Google NotebookLM Features Multi-format file upload: supports uploading documents in various formats, including PD...

Google DeepMind's Veo model, an advanced video generation model with extreme realism and fidelity, support for 4K output, and improved cue-following capabilities and creative control.Veo version 3 adds native audio generation, capable of generating sound effects, ambient soundscapes, and even dialog.Veo is an advanced video generation model that helps creatives make it easier to produce high-quality videos. It supports features such as native audio generation, style matching, and precise camera control for a variety of scenarios, from movie production to game development. v...

The Google I/O 2025 conference was officially held early this morning. At the conference, Google announced Gemini 2.5 Pro/Flash model, Imagen 4, Android XR glasses and other important products. Multimodal AI assistant "Project Astra": Google brought a multimodal AI assistant - Project Astra. Google wants to make Gemini a "world model" that can plan, understand, and model...

Google Inc. made an announcement yesterday (March 14), announcing the retirement of the classic version of Google Assistant, plans to gradually migrate users to Gemini in the coming months, and plans to stop the classic version of the Google Assistant service on mainstream mobile devices before the end of 2025 (old devices can still continue to use). Note: Google Assistant was launched in 2016 with the first generation of Pixel phones, with natural language processing and voice recognition technology, users can control their smart home by voice...

Freepik, a world-renowned creative resource platform, announced that it has partnered with tech giant Google to take the lead in launching the latest generation of AI video models - Veo 2 - globally. This heavyweight news quickly triggered widespread attention, and many users have said that this is probably the most advanced AI video generation tool at present. According to reports, Veo2 is developed by the Google DeepMind team and is a fully upgraded version of its predecessor model Veo. It can not only generate videos with up to 4K resolution, but also realize screen content up to several minutes long. What's more amazing is that Veo2 has a better sense of realism and animation smoothness in...

Tim Brooks, research lead for OpenAI's Sora video generation system, recently announced on social media that he will be joining Google DeepMind to continue his work on video generation and world simulators. Brooks said he was excited to work with such a talented team at DeepMind, and thanked OpenAI for the two great years he spent there, especially on the Sora project. Since its release in February of this year, Sora has attracted...

GenType is a letter image generator. It is an experimental tool launched by Google Labs that uses AI technology to generate stylized letter images. Users only need to enter a prompt word to describe the letter style they want to generate, and GenType will generate the corresponding letter image based on the description. For example, you can enter the prompt word "letter A composed of white clouds", and GenType will generate a picture of the letter A composed of white clouds. This tool is very suitable for creative design and art projects. The image generation speed is very fast and supports saving as PNG…

Google recently announced a series of major updates to its AI assistant Gemini, aimed at improving the user experience and expanding its scope of application. The update covers performance improvements, the introduction of new features, and the expansion of the user base. The core of the update is the introduction of Gemini 1.5Flash, which is a comprehensive upgrade of the free version of Gemini. Version 1.5Flash has significant improvements in response speed, reasoning ability, and image understanding. In particular, Gemini's context window has been expanded from the previous 8K to 32K...

Google's artificial intelligence assistant Gemini officially entered the social media platform today, opening an official account on X (formerly Twitter) and publishing its first tweet. This move not only marks Gemini's official debut in the social media field, but also triggers a lot of speculation about Google's brand strategy. The newly opened account username is @GeminiApp, and the full name is "Google Gemini App". It is worth noting that the Gemini brand is deliberately preceded by "Google&q…

Google recently launched an innovative Gmail sidebar feature called Gemini smart assistant, which aims to provide users with a more efficient email processing experience through its AI model (Gemini1.5Pro). This new feature allows users to perform multiple tasks, including summarizing email threads, suggesting email replies, helping to compose emails, and extracting specific information from inboxes or Google Drive files. Gemini smart assistant is designed to simplify the email management process and enable users to handle mail more efficiently. Example…

Tempus AI is a diagnostics company that uses artificial intelligence (AI) to interpret medical tests, aiming to help doctors provide more accurate treatment options for patients. On Friday, the company debuted on the Nasdaq under the ticker “TEM”, and its stock price rose as much as 15% after the opening. Tempus AI issued 11.1 million shares at $37 per share on Thursday, at the high end of its initial target range of $35 to $37. The company raised $410 million through the IPO, with a total valuation of more than $6 billion. Although the stock price rose in early trading…

After collaborating with Google Cloud, Aria today announced that it has integrated Google's Gemini model, bringing a major upgrade to its AI browser. Aria, driven by its own Composer AI engine, is known for its support for more than 150 local LLM variants with more than 50 model families. Now, with the addition of Google's Gemini, this advanced, powerful and user-friendly LLM model will provide users with the latest information and excellent performance. This collaboration will also promote Aria's image generation and speech output...

At Google's I/O developer conference, Google announced that it is testing a new call monitoring feature to alert users that they are on a call with someone who may be trying to defraud them and encourage them to end the call. Google said the feature uses Gemini Nano technology, a simplified version of the company's Gemini large language model developed for Android devices, which can run locally and offline to look for language and other conversation patterns associated with fraud. When these suspicious signs appear in a call, the user...

Google released a technical report on the Gemini1.5Pro model, which introduced the model architecture, training data and infrastructure, long text evaluation, and general capability evaluation of Gemini1.5Pro. Gemini1.5Pro is a high-performance multimodal hybrid expert model that can process and analyze information from millions of tokens, including long documents and hours of video and audio content. Gemini1.5Pro demonstrated almost perfect memory recall capabilities in long-form information retrieval tasks, refreshing long-form document question answering, long video question answering, and automatic...