By mid-2025 China had become a global leader in open-source large language models (LLMs). According to Chinese state media, by July 2025 China accounted for 1,509 of the world’s ~3,755 publicly released LLMs, far more than any other country. This explosion reflects heavy state and industry investment in domestic AI, open licensing (often Apache- or MIT-style), and a strategic pivot by Chinese tech giants and startups toward publicly shared models. The result is a "revival" of open-source AI, with dozens of Chinese LLMs now available for download or use via Hugging Face, GitHub, or cloud APIs. These range from general-purpose foundation models dozens of billions of parameters in size to specialized chatbots and domain experts, many built on Mixture-of-Experts (MoE) architectures.
Moondream transforms the humble Raspberry Pi into a context-aware visual interpreter, capable of answering nuanced questions about images in plain English. This guide explores its potential for home automation, security analysis, and more.
This article discusses the potential shift away from traditional graphical user interfaces (GUIs) towards interaction with computers through AI agents and natural language processing. It argues that AI is eliminating the need for windows, menus, and clicks, allowing users to simply tell computers what they need.
Introducing Aeneas, the first AI model for contextualizing ancient inscriptions, designed to help historians better interpret, attribute, and restore fragmentary texts. It reasons across thousands of Latin inscriptions, retrieving textual and contextual parallels to aid in historical research.
The article discusses how agentic LLMs can help users overcome the learning curve of the command line interface (CLI) by automating tasks and providing guidance. It explores tools like ShellGPT and Auto-GPT that leverage LLMs to interpret natural language instructions and execute corresponding CLI commands. The author argues that this approach can make the CLI more accessible and powerful, even for those unfamiliar with its intricacies.
This paper introduces Arch-Router, a preference-aligned routing framework for large language models (LLMs). It addresses limitations in existing routing approaches by focusing on matching queries to user-defined preferences (domain and action types) rather than solely relying on benchmark performance. The framework includes a 1.5B parameter model, Arch-Router, and a data creation pipeline. Experiments demonstrate state-of-the-art results in matching queries with human preferences and improved adaptability.
This article details a step-by-step guide on building a knowledge graph from plain text using an LLM-powered pipeline. It covers concepts like Subject-Predicate-Object triples, text chunking, and LLM prompting to extract structured information.
This article provides a beginner-friendly explanation of attention mechanisms and transformer models, covering sequence-to-sequence modeling, the limitations of RNNs, the concept of attention, and how transformers address these limitations with self-attention and parallelization.
This article provides a comprehensive guide on the basics of BERT (Bidirectional Encoder Representations from Transformers) models. It covers the architecture, use cases, and practical implementations, helping readers understand how to leverage BERT for natural language processing tasks.
The article provides a comprehensive introduction to large language models (LLMs), explaining their purpose, how they function, and their applications. It covers various types of LLMs, including general-purpose and task-specific models, and discusses the distinction between closed-source and open-source LLMs. The article also explores the ethical considerations of building and using LLMs and the future possibilities for these models.
