Search AI Products and News

Recently, with the rapid development of large models and RAG technology, the value of structured data in intelligent systems has become increasingly prominent. Against this backdrop, how to accurately convert unstructured data such as document images and PDFs into structured data has become a key challenge that the industry urgently needs to address. In response to this situation, the PaddlePaddle team, leveraging its deep technical expertise and profound insights into user needs, has launched the new-generation document parsing tool - PP-StructureV3, providing an innovative solution for solving complex document parsing problems. Currently, many open-source solutions struggle in handling complex

This comprehensive review of the Solar AI development tool examines how it generates complete applications via natural language input, reducing 83% of development time. Explore Solar's core features, use cases, and comparisons with mainstream tools to determine if it can truly deliver on its 'light-speed development' promise.

Recently, OpenAI announced the launch of its latest o3-pro AI model, which is designed to provide more reliable and accurate responses for enterprises. As an advanced version of the o series inference models, o3-pro supports more software tool integrations and is particularly suitable for businesses and developers requiring high precision and accuracy. However, the response speed of o3-pro has slowed down compared to previous models. OpenAI stated that due to o3-pro's enhanced capabilities with richer tools, it needs extra time when handling complex problems.

On June 6, MemSET Intelligence officially launched its latest masterpiece - the MiniCPM4.0 series model. This series is hailed as "the most imaginative pocket-sized powerhouse" in history. Not only does it achieve a leap in edge-side performance but also sets a new benchmark in technological innovation. The MiniCPM4.0 series includes two heavyweight products: one is an 8B Lightning Sparse Edition, which has sparked an efficient storm with its innovative sparse architecture; another is a nimble and lightweight 0.5B version, known as the "strongest pocket-sized powerhouse". These two models excel in speed, efficiency, performance, and application.

On June 3rd, Mary Meeker, known as the 'Internet Queen', released her 340-page report 'Trends - Artificial Intelligence' on May 30th. The report runs throughout with the word 'unprecedented', emphasizing that the development, adoption, investment, and application speed of AI technology has far exceeded any previous technological revolution. Key points: The speed of AI transformation overturns historical understanding. In the report, Meeker pointed out that the explosive growth trend of AI user numbers, technology iteration, and capital investment presents an 'upward-right' curve.

Recently, tech giant NVIDIA, in collaboration with the Massachusetts Institute of Technology (MIT) and the University of Hong Kong, released a new framework called Fast-dLLM. This innovative framework aims to significantly increase the inference speed of diffusion models (Diffusion-based LLMs), up to 27.6 times, providing stronger technical support for AI applications. The challenges and opportunities of diffusion models make them strong competitors to traditional autoregressive models.

Recently, MiTA AI Search officially launched the new ultra-fast model, bringing users a more efficient and accurate search experience. The MiTA AI Search team successfully achieved a maximum response speed of 400 tokens/second on a single H800 GPU through kernel fusion technology on GPUs and dynamic compilation optimization strategies on CPUs, with most responses provided within 2 seconds.