DeepSeek吸血AI板块,现在是该抄底还是撤退?

Odaily星球日报Published on 2025-01-27Last updated on 2025-01-27

Abstract

“更强大的模型对Agent来说总是好事。”

原创 | Odaily星球日报(@OdailyChina
作者|Azuma(@azuma_eth

DeepSeek引发AI Agent板块大洗牌?现在是该抄底还是撤退?


AI Agent 板块今日继续加速回调,除去部分 DeepSeek 概念的 meme 之外,VIRTUAL、AI16Z、AIXBT、ARC 等头部 Framework/Launchpad/Agent 代币以及其他大大小小代币均录得两位数的跌幅。

下跌固然与市场整体转弱有关,但环顾市场讨论,也有部分投资者会将本轮 AI Agent 的集体走低归因于 DeepSeek 的爆火,认为 DeepSeek 作为新一代更经济的 LLM 将会对当前的 Crypto AI 格局造成冲击甚至洗牌。


我个人并不支持这一观点。


上述质疑的核心要点其实在于 —— LLM 的迭代发展是否会颠覆现有的 Agent?抛开 DeepSeek 当前的推理质量是否已优于 ChatGPT、Claude 的纷论不提,仅从经济可持续的角度假设 DeepSeek 有着更可期的迭代潜力,那么迭代后的 DeepSeek 是否会再造一批新的 Agent?让现存的这些 Agent 变成“昨日黄花”?


解答该问题,需要先弄明白 LLM 与 Agent 的关系。我们尝试用 DeepSeek 自己与另一大“国产之光” Kimi 来解答这一问题(讲真,单说这一问 Kimi 的质量似乎更好)。


DeepSeek:
AI Agent(人工智能代理)和 AI LMM(大型语言模型)的关系在于,LMM 可以作为 Agent 的核心组件之一,特别是在处理自然语言理解和生成任务时。AI Agent 是一个能够自主执行任务、与环境交互的智能系统,而 LMM 则为 Agent 提供了强大的语言处理能力,使其能够更好地理解和生成人类语言,从而在对话、推荐、分析等场景中发挥作用。简而言之,LMM 是 AI Agent 实现智能化语言交互的重要技术支撑。


Kimi:

  • Agent 代理通常依赖 LLM 来理解和生成自然语言。LLM 提供了强大的语言处理能力,使得 Agent 能够与用户进行自然语言交互,理解用户的需求,并生成相应的响应。
  • Agent 也不仅仅依赖 LLM 的语言处理能力,还结合了决策、问题解决、与外部环境交互及执行动作等多方面的能力。LLM 为 Agent 提供了理解和生成语言的基础,而 Agent 则在此基础上进行任务规划和执行。
  • Agent 还可以结合其他工具和 API 来扩展其功能。例如,Agent 可以调用搜索引擎、数据库查询工具、文件读写工具等,以完成更复杂的任务。LLM 在这个过程中提供了语言理解和生成的能力,使得 AI 代理能够更好地与这些工具进行交互。


总结一下,Agent 的运行需要 LLM 的支持,但其表达与交互的质量并不完全取决于 LLM。实际上,决定了不同 Agent 呈现出明显差异的正是 LLM 之外的其他能力。


举个例子,aixbt 之所以能够在输出方面“碾压”其他同类型 Agent,本质上是因为其在提示词设计、后处理机制、上下文管理、微调策略、随机性控制、外部工具集成以及用户反馈机制等方面做得更好,所以能够生成更贴近于行业的话术表达 —— 你把它叫作先发优势也好,护城河也好,这就是 aixbt 当下的优势所在。 


在理解了这一关系逻辑之后,现在再来回答前文中的核心问题“LLM 的迭代发展是否会颠覆现有的 Agent”?

答案是否定的,因为 Agent 可以轻松地通过 API 集成来继承新一代 LLM 的能力来实现进化,从而提升交互质量提升,改善效率,扩展应用场景……尤其是考虑到 DeepSeek 本身就提供了与 OpenAI 兼容 API 格式。

事实上,反应够快的 Agent 早已完成了对 DeepSeek 的集成。ai16z 创始人 Shaw 今晨就曾表示,ai16z DAO 所开发的 AI Agent 构建框架 Eliza 已于两周前完成了对 DeepSeek 的支持。

DeepSeek引发AI Agent板块大洗牌?现在是该抄底还是撤退?


在当前的趋势下,我们可以理性假设继 ai16z 的 Eliza 之后,其他各大框架及 Agent 也会尽快完成对 DeepSeek 的集成。如此一来,即便短期内会遭遇某些新一代 DeepSeek Agent 的冲击,但长期来看各个 Agent 之间的竞争依旧取决于前文提到的外部能力,而这时先发优势所带来的开发成果积累将再度显现。


最后,还是放一些大佬针对 DeepSeek 的评论来为 AI Agent 板块的坚守者们充值一波信仰吧。


DeGods 创始人 Frank 昨日表示:“人们对这件事(DeepSeek 迭代旧市场)的想法是错误的,当前的 AI 项目将受益于像 DeepSeek 这样的新模型,他们只需将 OpenAI API 调用替换为 DeepSeek,一夜之间输出就会得到改善。新模型不会扰乱 Agent,而是会加速它们的发展。


专注于 AI 板块的交易员 Daniele 则表示:“如果你正在因 DeepSeek 模型价格低廉且开源而出售 AI 代币,那么你需要知道,DeepSeek 实际上非常有助于以低门槛定价将 AI 应用扩展到数百万用户。这可能是该行业迄今为止最好的事情。”


Shaw 今日上午也曾发布长文回应了 DeepSeek 的冲击,开篇第一句话如下:“更强大的模型对 Agent 来说总是好事。多年来,各大 AI 实验室一直在互相超越。有时 Google 领先,有时是 OpenAI,有时又是 Claude,今天则到了 DeepSeek……

Trending Cryptos

Related Reads

Unexpected Weak Non-Farm Payrolls Data Pushes BTC to Rebound 11%, FOMC Minutes to Test the Narrative of This Rally

Bitcoin has rebounded 11% from its 21-month low, but the sustainability of this rally hinges entirely on the Federal Reserve's release of the June FOMC meeting minutes. The bounce was triggered by a weaker-than-expected US jobs report, which showed only 57,000 jobs added in June—about half of economists' forecasts. This data prompted traders to scale back bets on further Fed rate hikes, fueling a rally in Bitcoin alongside gold and stocks. The upcoming minutes are critical. They will reveal whether Fed officials, in their mid-June meeting, were already expressing concerns about a weakening labor market, tight credit conditions, or the risks of overtightening—factors that would support the market's recent dovish shift. Conversely, if the discussion focused on persistent inflation and the conditions for more rate hikes, the rally's foundational narrative would crumble. Market indicators show the rebound's fragility. While US spot Bitcoin ETFs saw a significant single-day inflow, it followed a prolonged period of outflows. On-chain data indicates a substantial increase in Bitcoin being moved to exchanges, creating potential sell pressure. Options market positioning suggests key price levels around $60,000 and $62,000 that could either stabilize or accelerate price movement. In essence, Bitcoin's 11% gain is built on speculation about the Fed's private deliberations three weeks ago. The FOMC minutes will replace that speculation with concrete details, and the discrepancy between market expectations and the actual record will determine whether Bitcoin holds above $64,000 or falls back toward $58,000.

marsbit8m ago

Unexpected Weak Non-Farm Payrolls Data Pushes BTC to Rebound 11%, FOMC Minutes to Test the Narrative of This Rally

marsbit8m ago

Just Now, The World's First Ultra-High-Frame World Model Was Born, Nvidia Content 0, Racing to 50 FPS

Just Now, Global First Ultra-High-Frame World Model Born, 0% NVIDIA, Speeds to 50 FPS A Chinese team has developed MoWorld, the world's first Flash World Model, achieving real-time interactive inference exceeding 50 FPS. Crucially, it is entirely built on domestic NPUs (National Processing Units), bypassing NVIDIA GPUs. Developed by Moxin Technology in collaboration with Zhejiang University's Pan Yunhe academician team, MoWorld represents a complete, closed-loop system from training and distillation to deployment on domestic computing power. The model tackles the critical industry bottleneck of real-time performance, essential for applications like robotics, gaming, and digital worlds. MoWorld achieves this through a full-stack redesign for NPUs, including a proprietary 3D-annotated data pipeline, system-level optimizations for long-sequence training (up to 2000 frames), and inference optimizations like dynamic mixed-precision quantization. On a Huawei Ascend 910C platform, a 14B MoE parameter model achieves over 50 FPS, reducing typical inference costs by 70% compared to equivalent GPU solutions. This breakthrough lowers the deployment barrier, potentially accelerating the industrialization of world models. Key application areas include gaming/entertainment (offering 6-DoF camera control for immersive exploration), embodied AI/autonomous driving (providing a high-fidelity digital training ground), film pre-visualization, and 3D reconstruction/digital twins due to its strong geometric consistency. MoWorld demonstrates that a full-stack domestic compute ecosystem can support cutting-edge, real-time world models, positioning China at a competitive starting line in defining next-generation spatial intelligence standards. The project underscores a shift in competition from model scale to real-world usability and cost-effective deployment.

marsbit46m ago

Just Now, The World's First Ultra-High-Frame World Model Was Born, Nvidia Content 0, Racing to 50 FPS

marsbit46m ago

Pacific 'Fever': How Extreme Weather Becomes Wall Street's Cash Machine?

"Pacific Fever": How Extreme Weather Becomes Wall Street's ATM? The summer of 2026 sees unusually fierce weather across China and globally. The common driver behind this global pattern is a powerful El Niño event, potentially the strongest since 1950, as declared by NOAA. This phenomenon, characterized by warming central/eastern Pacific waters, disrupts global atmospheric circulation, raising risks of floods, droughts, and heatwaves, further intensified by climate change. For financial markets, especially commodities, El Niño is not just weather but a major trading theme. History shows its price impact is profound. In the 1970s, El Niño-driven anchovy collapse in Peru fueled a soybean boom, giving Richard Dennis his first million. Anthony Ward's cocoa empire was built on superior weather intelligence. Most recently in 2024, West African droughts caused cocoa prices to soar over 400%, delivering huge gains for trend-following hedge funds. In 2026, markets are again pricing in future El Niño-induced supply shocks. Despite high current inventories, prices for palm oil, rubber, and sugar have rallied on anticipation of upcoming Southeast Asian droughts and weak Indian monsoons. Analysts identify key indicators to watch: the Niño3.4 index, Indian monsoon rainfall, Malaysian palm oil stocks, and the fundraising scale of specialized weather funds like Moreton Capital. Beyond trading opportunities, a concerning narrative is gaining traction online, linking El Niño with fertilizer shortages and energy supply disruptions to warn of potential global food crises within months. While alarmist, it highlights a deeper truth: the cascading effects of climate-driven weather extremes ultimately translate into higher costs of living for everyone, far beyond the trading floor.

marsbit50m ago

Pacific 'Fever': How Extreme Weather Becomes Wall Street's Cash Machine?

marsbit50m ago

Pacific 'Fever': How Extreme Weather Becomes Wall Street's ATM?

"Pacific 'Fever': How Extreme Weather Becomes Wall Street's Piggy Bank" The article examines how the 2026-2027 El Niño, potentially the strongest since 1950, is not only disrupting global weather but also creating major financial opportunities. It links recent extreme events in China and worldwide to this climate phenomenon, which alters atmospheric patterns, increasing risks of floods, droughts, and heatwaves. The core narrative explores how financial markets capitalize on these disruptions. A hedge fund is raising $500 million specifically to bet on El Niño-affected crops like South African corn and Malaysian palm oil. Historically, such strategies have yielded massive profits. Examples include Richard Dennis ("Turtle Trader") making his first fortune in the 1970s soy boom triggered by El Niño's impact on Peruvian anchovies (a key fishmeal source), and Anthony Ward's cocoa empire built on superior weather intelligence. The 2024 cocoa price surge, driven by West African drought, enriched quantitative trend-following funds. Currently, markets are preemptively bidding up palm oil, rubber, and sugar futures based on anticipated future supply shocks, despite high current inventories. The article details El Niño's asymmetric global impacts: causing drought in Southeast Asia (hurting palm oil/rubber) and India (affecting sugar/cotton), but bringing beneficial rains to South American soy and sugarcane. Key metrics to watch include the Niño3.4 index, Indian monsoon data, and Malaysian palm oil stocks. The true price effects often materialize *after* the El Niño peaks, suggesting 2027 may see the real volatility. The conclusion warns that beyond trading gains, the convergence of El Niño, energy shortages, and fertilizer scarcity poses a systemic risk, potentially raising the cost of living for everyone, turning a climate event into a global economic story.

链捕手59m ago

Pacific 'Fever': How Extreme Weather Becomes Wall Street's ATM?

链捕手59m ago

Just Now, OpenAI's Chief Futurist Departed, Once Called a Jackass by Musk

Just now, OpenAI's Chief Futurist, Joshua Achiam, announced his departure from the company via X. Having joined as a 25-year-old intern in 2017, he spent nine years at OpenAI, evolving from an AI safety research scientist to leading the Mission Alignment team. Earlier this year, that team was dissolved, and Achiam transitioned to the newly created role of Chief Futurist, positioned at the intersection of AI safety and policy to study AGI's long-term risks and opportunities. In his departure statement, Achiam called his time a "graduation," reflecting on the immense progress from AI that couldn't converse to systems solving scientific problems. He expressed optimism about a future of peace, prosperity, and possibility, closing with "To safe AGI." His tenure was notably marked by a 2018 incident where he publicly challenged Elon Musk—then still with OpenAI—on safety compromises if Musk pursued AGI at Tesla, leading Musk to call him a "jackass." This became an internal legend, with colleagues later giving him a trophy inscribed, "To safety, never stop being that jackass." Achiam's exit follows a pattern of prominent safety and alignment experts leaving OpenAI, including Jan Leike and others who joined rivals like Anthropic or started non-profits. His departure coincides with OpenAI's internal efforts to more tightly integrate its research and policy teams, and the recent hiring of former White House AI advisor Dean Ball. Achiam did not cite a specific reason for leaving but indicated it was a long-considered decision, stating the mission to ensure AGI benefits humanity can now be advanced beyond the "frontier lab's" walls.

marsbit1h ago

Just Now, OpenAI's Chief Futurist Departed, Once Called a Jackass by Musk

marsbit1h ago

Trading

Spot

Hot Articles

How to Buy VIRTUAL

Welcome to HTX.com! We've made purchasing Virtuals Protocol (VIRTUAL) simple and convenient. Follow our step-by-step guide to embark on your crypto journey.Step 1: Create Your HTX AccountUse your email or phone number to sign up for a free account on HTX. Experience a hassle-free registration journey and unlock all features.Get My AccountStep 2: Go to Buy Crypto and Choose Your Payment MethodCredit/Debit Card: Use your Visa or Mastercard to buy Virtuals Protocol (VIRTUAL) instantly.Balance: Use funds from your HTX account balance to trade seamlessly.Third Parties: We've added popular payment methods such as Google Pay and Apple Pay to enhance convenience.P2P: Trade directly with other users on HTX.Over-the-Counter (OTC): We offer tailor-made services and competitive exchange rates for traders.Step 3: Store Your Virtuals Protocol (VIRTUAL)After purchasing your Virtuals Protocol (VIRTUAL), store it in your HTX account. Alternatively, you can send it elsewhere via blockchain transfer or use it to trade other cryptocurrencies.Step 4: Trade Virtuals Protocol (VIRTUAL)Easily trade Virtuals Protocol (VIRTUAL) on HTX's spot market. Simply access your account, select your trading pair, execute your trades, and monitor in real-time. We offer a user-friendly experience for both beginners and seasoned traders.

6.1k Total ViewsPublished 2024.12.23Updated 2026.06.02

How to Buy VIRTUAL

Discussions

Welcome to the HTX Community. Here, you can stay informed about the latest platform developments and gain access to professional market insights. Users' opinions on the price of VIRTUAL (VIRTUAL) are presented below.

活动图片