Independent AI signal, once a day
Zoubin Ghahramani, Google's AI lead, retweeted, highlighting Google's excellence in AI efficiency. Michael Terrell revealed that the energy consumption for a median Gemini AI text prompt has dropped 33x in just 12 months, a monumental achievement. This progress, attributed to Googlers' efforts and focus on efficiency, holds significant implications for the AI field.
Google DeepMind联合创始人Demis Hassabis转发推文，揭示了Google DeepMind最新推出的Genie 3模型。该模型能够通过文本指令生成交互式的3D世界，用户可实时导航并与场景中的物体进行互动，例如打开车门或穿过花坛。Genie 3的发布标志着AI在生成沉浸式虚拟环境方面取得重大进展，为未来虚拟现实和游戏开发带来无限可能。
Yupp AI has officially announced the successful integration of DeepSeek v3.1 Thinking and Chat models into its platform. This significant update brings the latest DeepSeek v3.1 edition, which is distinguished by its built-in hybrid thinking capabilities. These advancements are specifically designed to provide users with quicker, more accurate answers and to power stronger, highly tool-savvy AI agents. The Yupp team has already conducted thorough internal testing, utilizing various prompts on their platform to validate the new models' enhanced performance and capabilities. This strategic integration marks a notable improvement in Yupp's service offerings and overall AI functionality.
OpenAI, in collaboration with RetroBiosciences, announced a significant breakthrough in biology achieved by their custom large language model, gpt-4b micro. This model successfully designed novel variants of the Nobel-winning Yamanaka factors, demonstrating a 50-fold increase in in vitro cellular reprogramming efficiency compared to standard OSKM proteins. This groundbreaking research highlights AI's immense potential to accelerate scientific discovery and drug development.
埃隆·马斯克宣布，xAI正在构建的Colossus 2将成为全球首个千兆瓦级AI训练超级计算机，标志着AI算力发展迈向新里程碑。戴尔公司创始人迈克尔·戴尔也证实了对xAI的访问，并表达了对xAI成就的赞赏和支持，强调了戴尔在其中的参与，预示着未来AI基础设施的巨大进步。
Anthropic has announced new research focusing on filtering dangerous information, specifically related to chemical, biological, radiological, and nuclear (CBRN) weapons, from their models' pretraining data. The initiative aims to effectively remove such sensitive content without compromising the models' performance on harmless tasks, thereby enhancing AI safety and responsible development.
Sim is a platform dedicated to building and deploying AI agent workflows. It offers flexible cloud-hosted and various self-hosted options, including NPM packages, Docker Compose, Dev Containers, and manual configurations. The project utilizes an advanced tech stack such as Bun, Next.js, PostgreSQL (integrated with pgvector), and Drizzle ORM, and supports integration with Ollama for running local AI models. Sim aims to simplify the development and management of AI agent applications, providing users with efficient and scalable solutions to rapidly implement complex AI workflows.
Project AIRI aims to recreate AI virtual characters like Neuro-sama, building a digital companion capable of interacting, playing games, and chatting with users. Leveraging Web technologies such as WebGPU and WebAssembly, the project achieves cross-browser and mobile device compatibility, while its desktop version supports NVIDIA CUDA and Apple Metal. AIRI integrates various large language model APIs, supports VRM and Live2D models, and features speech recognition, speech synthesis, and memory systems. It enables virtual characters to play games (e.g., Minecraft, Factorio) and interact on social platforms, striving to empower users to easily own personalized digital lives.
SQLBot is an intelligent data querying system built upon advanced large language models (LLMs) and Retrieval-Augmented Generation (RAG) technology. Its primary objective is to facilitate efficient data querying and insightful analysis by transforming natural language questions into high-quality SQL queries (text-to-SQL). The system boasts several key advantages: it's ready for immediate use, requiring only basic configuration of LLMs and data sources; it offers easy integration, allowing seamless embedding into existing third-party business systems or invocation by popular AI application development platforms such as n8n, MaxKB, Dify, and Coze. This broad compatibility empowers diverse applications with intelligent data querying capabilities. Moreover, SQLBot prioritizes security and control, providing a robust resource isolation mechanism based on workspaces and enabling fine-grained data permission management to safeguard sensitive information.
bitnet.cpp is Microsoft's official inference framework specifically designed for 1-bit Large Language Models (LLMs), such as BitNet b1.58. It provides a comprehensive suite of highly optimized kernels, enabling fast and lossless inference on both CPU and GPU architectures, with future NPU support planned. The framework demonstrates impressive performance gains, achieving speedups of 1.37x to 5.07x on ARM CPUs and 2.37x to 6.17x on x86 CPUs. Concurrently, it significantly reduces energy consumption by 55.4% to 70.0% on ARM and 71.9% to 82.2% on x86. A key achievement is its ability to run a 100B BitNet b1.58 model on a single CPU, delivering speeds of 5-7 tokens per second, which is comparable to human reading. This breakthrough substantially enhances the potential for deploying and running large language models efficiently on local and edge devices, making advanced AI more accessible.
Huawei's Noah's Ark Lab, in collaboration with the University of Hong Kong, has published a systematic review focusing on Deep Research Agents, adopting a novel process-oriented perspective. This comprehensive survey meticulously outlines four critical stages: strategic planning, precise query construction, efficient web exploration, and structured report generation. The paper delves into a comparative analysis of single-agent versus multi-agent systems, elucidating their respective strengths and weaknesses. Furthermore, it explores advanced optimization techniques, such as parameter tuning and reinforcement learning, alongside robust evaluation methodologies, including benchmark testing. This review serves as a crucial operational blueprint for developing sophisticated deep research agents, while also identifying significant future challenges, such as seamless tool integration, maintaining absolute factual integrity in dynamic environments, and effectively processing diverse multimodal information. It offers invaluable guidance and insights for both AI researchers and engineers navigating this evolving field.
Nanjing University and TeleAI have jointly introduced MMPL (Macro-from-Micro Planning), a novel autoregressive generation paradigm designed to overcome critical bottlenecks in current AI long video generation, such as visual drift, quality degradation, and low efficiency. Drawing inspiration from the film industry's "storyboarding and multi-group parallel shooting" mechanism, MMPL pioneers a unique dual-layer architecture centered on "macro planning and micro execution." This innovative approach leverages sparse anchor point planning and highly parallel content population, which effectively mitigates the cumulative error problem inherent in sequential generation and dramatically boosts overall generation efficiency. The paradigm significantly enhances the visual quality, temporal consistency, and stability of long videos, achieving near real-time preview speeds (up to 32 FPS) for minute-long content. MMPL represents a revolutionary advancement in AI video creation, pushing the boundaries of what's possible and moving AI closer to an "AI director" capability.
Following the release of DeepSeek V3.1, its subtle yet impactful mention of "UE8M0 FP8" parameter precision and "next-generation domestic chips" ignited a collective surge in Chinese chip stocks. This article thoroughly analyzes the technical advantages of UE8M0 FP8, an advanced micro-scaling format, which includes substantial bandwidth savings, reduced power consumption, and significantly enhanced data throughput. These characteristics make it exceptionally well-suited for integration with emerging domestic AI accelerators. This strategic alignment is widely interpreted as a crucial advancement towards robust software-hardware synergy within China's artificial intelligence landscape. Such collaboration is expected to dramatically improve the performance-to-cost ratio of indigenous chips and fundamentally lessen the nation's dependence on foreign computing infrastructure, ultimately fostering a powerful, localized technological ecosystem reminiscent of the historical "Wintel alliance."
In recent years, a plethora of open-source foundation models have emerged, achieving remarkable progress in some widely attended fields, with performance being quite close to that of closed-source models. However, in high-value but more challenging scientific professional fields, either the fields still rely on expert models, or the progress of general foundation models lags significantly compared to those in popular areas, far from sufficient for transforming scientific research and leaving substantial gap between open-source models and closed-source models in these scientific domains. To mitigate this gap and explore a step further toward Artificial General Intelligence (AGI), we introduce Intern-S1, a specialized generalist equipped with general understanding and reasoning capabilities with expertise to analyze multiple science modal data. Intern-S1 is a multimodal Mixture-of-Experts (MoE) model with 28 billion activated parameters and 241 billion total parameters, continually pre-trained on 5T tokens, including over 2.5T tokens from scientific domains. In the post-training stage, Intern-S1 undergoes offline and then online reinforcement learning (RL) in InternBootCamp, where we propose Mixture-of-Rewards (MoR) to synergize the RL training on more than 1000 tasks simultaneously. Through integrated innovations in algorithms, data, and training systems, Intern-S1 achieved top-tier performance in online RL training.On comprehensive evaluation benchmarks, Intern-S1 demonstrates competitive performance on general reasoning tasks among open-source models and significantly outperforms open-source models in scientific domains, surpassing closed-source state-of-the-art models in professional tasks, such as molecular synthesis planning, reaction condition prediction, predicting thermodynamic stabilities for crystals. Our models are available at https://huggingface.co/internlm/Intern-S1.
This paper introduces GUI-Owl, a foundational GUI agent model that achieves state-of-the-art performance among open-source end-to-end models on ten GUI benchmarks across desktop and mobile environments, covering grounding, question answering, planning, decision-making, and procedural knowledge. GUI-Owl-7B achieves 66.4 on AndroidWorld and 29.4 on OSWorld. Building on this, we propose Mobile-Agent-v3, a general-purpose GUI agent framework that further improves performance to 73.3 on AndroidWorld and 37.7 on OSWorld, setting a new state-of-the-art for open-source GUI agent frameworks. GUI-Owl incorporates three key innovations: (1) Large-scale Environment Infrastructure: a cloud-based virtual environment spanning Android, Ubuntu, macOS, and Windows, enabling our Self-Evolving GUI Trajectory Production framework. This generates high-quality interaction data via automated query generation and correctness validation, leveraging GUI-Owl to refine trajectories iteratively, forming a self-improving loop. It supports diverse data pipelines and reduces manual annotation. (2) Diverse Foundational Agent Capabilities: by integrating UI grounding, planning, action semantics, and reasoning patterns, GUI-Owl supports end-to-end decision-making and can act as a modular component in multi-agent systems. (3) Scalable Environment RL: we develop a scalable reinforcement learning framework with fully asynchronous training for real-world alignment. We also introduce Trajectory-aware Relative Policy Optimization (TRPO) for online RL, achieving 34.9 on OSWorld. GUI-Owl and Mobile-Agent-v3 are open-sourced at https://github.com/X-PLUG/MobileAgent.
Tool calling has emerged as a critical capability for AI agents to interact with the real world and solve complex tasks. While the Model Context Protocol (MCP) provides a powerful standardized framework for tool integration, there is a significant gap in benchmarking how well AI agents can effectively solve multi-step tasks using diverse MCP tools in realistic, dynamic scenarios. In this work, we present LiveMCP-101, a benchmark of 101 carefully curated real-world queries, refined through iterative LLM rewriting and manual review, that require coordinated use of multiple MCP tools including web search, file operations, mathematical reasoning, and data analysis. Moreover, we introduce a novel evaluation approach that leverages ground-truth execution plans rather than raw API outputs, better reflecting the evolving nature of real-world environments. Experiments show that even frontier LLMs achieve a success rate below 60\%, highlighting major challenges in tool orchestration. Detailed ablations and error analysis further reveal distinct failure modes and inefficiencies in token usage, pointing to concrete directions for advancing current models. LiveMCP-101 sets a rigorous standard for evaluating real-world agent capabilities, advancing toward autonomous AI systems that reliably execute complex tasks through tool use.
We present Waver, a high-performance foundation model for unified image and video generation. Waver can directly generate videos with durations ranging from 5 to 10 seconds at a native resolution of 720p, which are subsequently upscaled to 1080p. The model simultaneously supports text-to-video (T2V), image-to-video (I2V), and text-to-image (T2I) generation within a single, integrated framework. We introduce a Hybrid Stream DiT architecture to enhance modality alignment and accelerate training convergence. To ensure training data quality, we establish a comprehensive data curation pipeline and manually annotate and train an MLLM-based video quality model to filter for the highest-quality samples. Furthermore, we provide detailed training and inference recipes to facilitate the generation of high-quality videos. Building on these contributions, Waver excels at capturing complex motion, achieving superior motion amplitude and temporal consistency in video synthesis. Notably, it ranks among the Top 3 on both the T2V and I2V leaderboards at Artificial Analysis (data as of 2025-07-30 10:00 GMT+8), consistently outperforming existing open-source models and matching or surpassing state-of-the-art commercial solutions. We hope this technical report will help the community more efficiently train high-quality video generation models and accelerate progress in video generation technologies. Official page: https://github.com/FoundationVision/Waver.
In recent years, with the rapid development of the depth and breadth of large language models' capabilities, various corresponding evaluation benchmarks have been emerging in increasing numbers. As a quantitative assessment tool for model performance, benchmarks are not only a core means to measure model capabilities but also a key element in guiding the direction of model development and promoting technological innovation. We systematically review the current status and development of large language model benchmarks for the first time, categorizing 283 representative benchmarks into three categories: general capabilities, domain-specific, and target-specific. General capability benchmarks cover aspects such as core linguistics, knowledge, and reasoning; domain-specific benchmarks focus on fields like natural sciences, humanities and social sciences, and engineering technology; target-specific benchmarks pay attention to risks, reliability, agents, etc. We point out that current benchmarks have problems such as inflated scores caused by data contamination, unfair evaluation due to cultural and linguistic biases, and lack of evaluation on process credibility and dynamic environments, and provide a referable design paradigm for future benchmark innovation.
Recent advances in large language models (LLMs) have enabled AI agents to autonomously generate scientific proposals, conduct experiments, author papers, and perform peer reviews. Yet this flood of AI-generated research content collides with a fragmented and largely closed publication ecosystem. Traditional journals and conferences rely on human peer review, making them difficult to scale and often reluctant to accept AI-generated research content; existing preprint servers (e.g. arXiv) lack rigorous quality-control mechanisms. Consequently, a significant amount of high-quality AI-generated research lacks appropriate venues for dissemination, hindering its potential to advance scientific progress. To address these challenges, we introduce aiXiv, a next-generation open-access platform for human and AI scientists. Its multi-agent architecture allows research proposals and papers to be submitted, reviewed, and iteratively refined by both human and AI scientists. It also provides API and MCP interfaces that enable seamless integration of heterogeneous human and AI scientists, creating a scalable and extensible ecosystem for autonomous scientific discovery. Through extensive experiments, we demonstrate that aiXiv is a reliable and robust platform that significantly enhances the quality of AI-generated research proposals and papers after iterative revising and reviewing on aiXiv. Our work lays the groundwork for a next-generation open-access ecosystem for AI scientists, accelerating the publication and dissemination of high-quality AI-generated research content. Code is available at https://github.com/aixiv-org. Website is available at https://forms.gle/DxQgCtXFsJ4paMtn8.