# LLM İlgili Makaleler

HTX Haber Merkezi, kripto endüstrisindeki piyasa trendleri, proje güncellemeleri, teknoloji gelişmeleri ve düzenleyici politikaları kapsayan "LLM" hakkında en son makaleleri ve derinlemesine analizleri sunmaktadır.

Conversation with Mai-Lan from AWS: The Next Battlefield for S3 – How to Handle the Data Consumption Surge in the Agent Era

The explosive rise of Agent AI, exemplified by OpenClaw in China, is putting unprecedented pressure on cloud data infrastructure. Unlike human engineers, Agents consume data in an "extremely active and aggressive" parallel fashion, launching tens to hundreds of queries simultaneously, leading to exponentially higher call frequencies and throughput. Mai-Lan Tomsen Bukovec, VP of Technology at AWS, emphasizes that cost-effectiveness in this data layer is now a decisive factor for customers building Agent systems. To address this, AWS is positioning its foundational Amazon S3 service, now 20 years old, as the critical data platform for the Agent era. Recent key innovations include: **S3 Table** with native Apache Iceberg support, enabling Agents to efficiently interact with structured data via familiar SQL; **S3 Vector**, which introduces vectors as a native type for building contextual data and serving as a shared "memory space" for AI systems; and the newly launched **S3 Files**, which provides a POSIX-compliant file system interface over S3, allowing Agents to interact with data through the familiar paradigm of files and directories. These enhancements are designed to meet the unique data interaction patterns of Agents, which are trained on models already proficient with SQL, file systems, and contextual vectors. By unifying these access methods on the scalable, durable, and cost-efficient S3 foundation, AWS aims to provide the data backbone capable of supporting the next wave of hyper-scale, high-frequency Agent applications.

marsbit12 saat önce

Conversation with Mai-Lan from AWS: The Next Battlefield for S3 – How to Handle the Data Consumption Surge in the Agent Era

marsbit12 saat önce

How Many Tokens Away Is Yang Zhilin from the 'Moon Chasing the Light'?

The article explores the intense competition between two leading Chinese AI companies, DeepSeek and Kimi (Moon Dark Side), and the mounting pressure on Yang Zhilin, the founder of Kimi. While DeepSeek re-emerged after 15 months of silence with its powerful V4 model—boasting 1.6 trillion parameters and low-cost, long-context capabilities—Kimi has been focusing on long-context processing and multi-agent systems with its K2.6 model. Yang faces a threefold challenge: technological rivalry, commercialization pressure, and investor expectations. Despite Kimi’s high valuation (reaching $18 billion), its revenue heavily relies on a single product with low paid conversion rates, while DeepSeek’s strategic silence and open-source influence have strengthened its market position and valuation prospects, now targeting over $20 billion. Both companies reflect broader trends in China’s AI ecosystem: Kimi aims for global influence through open-source contributions and agent-based advancements, while DeepSeek prioritizes foundational innovation and hardware independence, notably shifting to Huawei’s chips. Their competition is seen as vital for China’s AI progress, with the gap between top Chinese and U.S. models narrowing to just 2.7% on the Elo rating scale. Ultimately, the article argues that this rivalry, though anxiety-inducing for leaders like Zhilin, is essential for driving innovation and solidifying China’s role in the global AI landscape.

marsbit04/26 11:25

How Many Tokens Away Is Yang Zhilin from the 'Moon Chasing the Light'?

marsbit04/26 11:25

Computing Power Constrained, Why Did DeepSeek-V4 Open Source?

DeepSeek-V4 has been released as a preview open-source model, featuring 1 million tokens of context length as a baseline capability—previously a premium feature locked behind enterprise paywalls by major overseas AI firms. The official announcement, however, openly acknowledges computational constraints, particularly limited service throughput for the high-end DeepSeek-V4-Pro version due to restricted high-end computing power. Rather than competing on pure scale, DeepSeek adopts a pragmatic approach that balances algorithmic innovation with hardware realities in China’s AI ecosystem. The V4-Pro model uses a highly sparse architecture with 1.6T total parameters but only activates 49B during inference. It performs strongly in agentic coding, knowledge-intensive tasks, and STEM reasoning, competing closely with top-tier closed models like Gemini Pro 3.1 and Claude Opus 4.6 in certain scenarios. A key strategic product is the Flash edition, with 284B total parameters but only 13B activated—making it cost-effective and accessible for mid- and low-tier hardware, including domestic AI chips from Huawei (Ascend), Cambricon, and Hygon. This design supports broader adoption across developers and SMEs while stimulating China's domestic semiconductor ecosystem. Despite facing talent outflow and intense competition in user traffic—with rivals like Doubao and Qianwen leading in monthly active users—DeepSeek has maintained technical momentum. The release also comes amid reports of a new funding round targeting a valuation exceeding $10 billion, potentially setting a new record in China’s LLM sector. Ultimately, DeepSeek-V4 represents a shift toward open yet realistic infrastructure development in the constrained compute landscape of Chinese AI, emphasizing engineering efficiency and domestic hardware compatibility over pure model scale.

marsbit04/26 00:27

Computing Power Constrained, Why Did DeepSeek-V4 Open Source?

marsbit04/26 00:27

DeepSeek No Longer Wants to Focus Only on Large Models

DeepSeek, a leading Chinese AI company, has released its new model series DeepSeek-V4, featuring two versions: the high-performance V4-Pro with 1.6 trillion parameters and the cost-efficient V4-Flash. Both support 1 million token context windows and use Mixture-of-Experts (MoE) architecture to improve efficiency. The company continues its strategy of offering competitive pricing, with input tokens priced as low as ¥0.2 per million tokens. A key revelation is DeepSeek’s explicit link between future price reductions and the mass availability of Huawei’s Ascend 950 AI chips in the second half of the year. This signals a strategic shift from relying solely on algorithmic and engineering optimizations to integrating domestic computing power into its core cost structure. DeepSeek has adapted its inference system to run efficiently on both NVIDIA GPUs and Huawei NPUs, potentially challenging NVIDIA's CUDA ecosystem dominance. Concurrently, DeepSeek is reportedly seeking significant external investment, with a pre-money valuation of around ¥300 billion. This move highlights growing pressures in scaling compute infrastructure, retaining top talent—amid recent departures of key researchers—and accelerating commercialization efforts. The company has also updated its consumer app with tiered model access, indicating a stronger product focus. The V4 release underscores that China's AI competition is evolving beyond pure model capability into a broader contest involving compute supply chains, engineering systems, financing, and talent strategy.

marsbit04/25 01:45

DeepSeek No Longer Wants to Focus Only on Large Models

marsbit04/25 01:45

活动图片