Search AI Products and News

Pruna AI, a European startup focused on AI model compression algorithms, recently announced the open-sourcing of its optimization framework to help developers more efficiently compress AI models. The framework developed by Pruna AI combines several efficiency methods, including caching, pruning, quantization, and distillation, aiming to improve AI model performance. It not only standardizes the saving and loading of compressed models but also evaluates the compressed models to determine whether their quality has significantly degraded, while simultaneously measuring performance.

On March 13th, the AI interaction platform Chat Playground received a major update and was officially renamed Prompts Playground. According to the official announcement, this upgrade not only involves a name change but also a complete redesign of the platform, aimed at improving user experience and helping users more efficiently test, compare, and iterate prompts. The new Prompts Playground introduces several useful tools, including Web search integration and more.

In the fiercely competitive AI landscape, a million-dollar experiment is quietly revolutionizing large language model (LLM) training. Jieyue Xingchen's research team recently released groundbreaking findings. By utilizing nearly 1 million NVIDIA H800 GPU hours, they trained 3,700 models of varying sizes from scratch, accumulating a staggering 100 trillion tokens. This revealed a universal scaling law dubbed 'Step Law,' offering a novel guide for efficient LLM training.

DeepSeek launched an official account on Zhihu and published a technical article, "DeepSeek-V3/R1 Inference System Overview," detailing its model inference system optimization and cost-profit margin information for the first time, marking the end of the highly anticipated DeepSeek Open Source Week. The article highlights two key optimization goals for the DeepSeek-V3/R1 inference system: higher throughput and lower latency. To achieve these goals, DeepSeek employs large-scale cross-node expert parallelism (EP) technology, although...

In 2024, with the rise of AI-generated search results, many websites are experiencing a sharp decline in organic traffic. In response to this trend, Austrian startup Otterly has developed a tool focused on AI Search Visibility Optimization to help businesses and advertising agencies enhance their presence in generative search results. With Google launching AI Overviews in May, search engine users can...

As artificial intelligence increasingly integrates into enterprise workflows and products, the demand for Machine Learning Operations (MLOps) platforms is rising. These platforms assist companies in creating, testing, and deploying machine learning models more easily. However, despite the presence of several competitors in the market, such as startups like InfuseAIComet and major companies like Google Cloud, Azure, and AWS, South Korea's VESSL AI aims to carve out its niche by focusing on optimizing GPU costs.

Japanese electric vertical takeoff and landing (eVTOL) car manufacturer SkyDrive recently announced that they are collaborating with Braid Technologies to leverage artificial intelligence for optimizing the design of their flying cars. With AI technology, SkyDrive can generate thousands of design patterns, helping to fine-tune the structure of their electric aircraft for greater efficiency. Image source note: Image generated by AI, image authorized by service provider Midjourney.

In the realm of printers, HP's recent innovative moves have sparked widespread attention. Indeed, they are not focused on improving frequent paper jam issues or considering how to make it easier for you to use affordable third-party ink. Instead, they are bringing artificial intelligence technology into printers, offering users a brand new printing experience. This new feature called 'HP Print AI' is currently in an exclusive testing phase, generating high expectations. Firstly, it introduces the 'Perfect Output' feature aimed at improving the printing quality of web pages. In the past, printing web content often...

The Transformer architecture, a star in the field of artificial intelligence, has led a revolution in natural language processing with its self-attention mechanism at its core. However, when handling long contexts, the resource consumption of self-attention calculations becomes a bottleneck. To address this issue, researchers have proposed the Tree Attention method, which decomposes the calculation tasks through tree reduction to improve efficiency. This method not only reduces communication overhead and memory usage but is also 8 times faster than existing methods in a multi-GPU environment. The introduction of Tree Attention not only enhances performance.

Google's text-to-image model Imagen3's 'ImageFX' feature is now officially available to all users, offering an enhanced prompt interaction experience. Users only need to input descriptive text, and the system can generate corresponding images while optimizing the prompts, providing various options to improve the generation results. Furthermore, Imagen3 also includes a localized modification feature, allowing users to use brush tools to edit specific parts of the image and fill in related descriptive terms, with the system automatically generating the modified image, greatly improving the flexibility and convenience of image generation.

Wix has launched a new generative AI tool that can automatically create SEO-optimized blog posts and images, available to English-speaking users. This tool can recommend topics based on past content, automatically gather information to generate specific posts, provide SEO keyword suggestions, and allow users to create images through descriptions. Business users can connect their blogs with the Wix business solutions platform to enjoy services such as promotional emails. While this tool offers convenience, it may raise concerns about the proliferation of AI-generated content online. Wix has a precedent of engaging with AI.