Search AI Products and News

FizzDragon, a Singaporean film production company, recently released the 70-minute Mandarin AIGC (AI-generated content) film 'Zheng Yi Sao, Queen of the Seas' in Singaporean cinemas. This marks the global premiere of an AI-generated feature film in commercial theaters. The film is based on the life of Zheng Yi Sao, a legendary 18th-century Chinese pirate queen, whose story reportedly inspired the character of 'Tia Dalma' in the Pirates of the Caribbean series.

A GitHub open-source project named "system-prompts-and-models-of-ai-tools" has gained significant traction, accumulating 30.5K stars and becoming a popular resource for AI developers and researchers. This project, according to AIbase, compiles system prompts and model configurations for 9 mainstream AI tools, encompassing over 6500 lines of content and covering v0, Cursor, Manus, Same.dev, Lovable, Devin, and Rep.

At the recent Shanghai Auto Show, Zhuoyu Technology (hereinafter referred to as "Zhuoyu") announced its successful integration of Alibaba's Tongyi large model, marking another significant advancement in the field of intelligent driving. As an intelligent driving solutions provider and service provider, Zhuoyu leverages Alibaba Cloud's powerful computing capabilities to build an end-to-end intelligent driving world model, aiming to enhance the intelligence level of vehicles. In recent years, intelligent driving technology has rapidly developed, with major automakers significantly increasing their investment in this area. Zhuoyu Technology's collaboration with Alibaba's Tongyi large model has...

Zhipu BigModel's open platform has adjusted prices for several of its model offerings. GLM-4-FlashX, for example, is now priced at just 10 RMB per 100 million tokens. Built on a powerful pre-trained base, this model boasts exceptionally fast inference speeds and functional capabilities comparable to GPT-4, excelling in data extraction, generation, and translation.

Mozilla recently launched a new feature in Firefox Labs – "Link Preview." This feature aims to improve browsing experience by allowing users to quickly understand the content of a link without opening a new page, using a simple mouse operation. With the feature enabled, users only need to hold down the Shift and Alt keys and hover their mouse over any link to have a preview card pop up. The card displays the page title, a short description, images, estimated reading time, and three automatically generated summary points.

On April 24th, Kunlun Wanwei announced the official open-sourcing of its multimodal reasoning model, Skywork-R1V2.0 (hereinafter referred to as R1V2.0). This upgraded version demonstrates significant improvements in both visual and text reasoning capabilities, particularly excelling in deep reasoning for challenging science problems in the College Entrance Examination and general task scenarios. It is considered the currently most balanced open-source multimodal model, equally adept at visual and text reasoning.

Researchers from Hong Kong and the UK have recently proposed a novel image tokenization method designed to convert images into digital representations (tokens) in a more compact and accurate manner. Unlike traditional methods that distribute information uniformly across all tokens, this method employs a hierarchical structure, capturing visual information layer by layer, thereby improving the quality and efficiency of image reconstruction. Traditional image tokenization techniques typically divide each part of an image equally into multiple tokens, whereas the new method adopts a hierarchical structure. Initial tokens encode rough shapes and structural elements,

OpenAI recently released its latest AI model, GPT-4.1, claiming superior instruction following. However, independent tests suggest a decline in alignment, i.e., reliability, compared to its predecessor, GPT-4. OpenAI typically releases detailed technical reports including safety evaluations with new models, but hasn't done so this time, explaining that GPT-4.1 is not considered a 'cutting-edge' model.

Israel is poised to become the first country globally to implement AI tutoring nationwide. Startup eSelf, in collaboration with the Center for Educational Technology (CET), Israel's largest K-12 textbook publisher, will provide personalized AI personal tutoring to students across the country. Harvard University will serve as an academic advisor for the pilot program, evaluating its educational framework and impact. The initial pilot phase, launching in May, is expected to involve 10,000 students. This collaboration aims to address two major challenges facing education today: declining student achievement and a global teacher shortage.

Ainos, the leader in AI-powered odor digitization, announces the successful integration of its proprietary AI Nose olfactory module into a humanoid robot developed by Japan's leading service robotics company, ugo, Inc. Completed on April 9, 2025, this installation marks the world's first robot with a functional sense of smell, ushering in a new era for robotics, AI, and human-robot interaction. Image Note: Image generated by AI, image licensing provider MidjourneyAinos' AI

Anthropic recently released a comprehensive best practices guide for Claude Code, providing developers with a low-level, command-line interface (CLI)-centric tool to seamlessly integrate the Claude large language model into their daily programming tasks. Based on Anthropic's internal best practices, this guide emphasizes flexible, secure, and efficient coding patterns, offering valuable guidance for engineers looking to incorporate AI into their existing development environments.