7*24

Hot topic

Market

Activities

According to monitoring by 1M AI News, the web and app services of AI large model company DeepSeek experienced performance issues again in the early hours of March 30. The official status page indicated that the team implemented a fix at 1:24 AM and entered a monitoring phase, but 52 minutes later (at 2:16 AM), the problem reoccurred, leading to a return to investigation status. As of the time of publication, the incident remains unresolved.

[DeepSeek Web/App Services Malfunction Again Less Than an Hour After Repair]According to monitoring by 1M AI News, the web and app services of AI large model company DeepSeek experienced performance issues again in the early hours of March 30. The official status page indicated that the team implemented a fix at 1:24 AM and entered a monitoring phase, but 52 minutes later (at 2:16 AM), the problem reoccurred, leading to a return to investigation status. As of the time of publication, the incident remains unresolved.

DeepSeek Web/App Services Malfunction Again Less Than an Hour After Repair

On March 30, crypto analyst Willy Woo stated that some traditional on-chain models suggest that the bottom for BTC is roughly between $46,000 and $54,000, while also indicating that a waiting period may still be necessary. The capital stored in BTC has been flowing out since November. However, it is important to note that these models have limitations. They are based on historical behavior, and Bitcoin has only experienced four bear markets in its history, all occurring against the backdrop of a long-term bull market for risk assets. If this foundation is broken, we may enter unknown territory (a deeper bear market). Ultimately, I believe that due to the potential collapse of the long-term bull market structure on a global macro level, the probability of a deeper bear market is quite high.

[Willy Woo: Traditional On-Chain Models Indicate Bitcoin Bottom Between $46,000 and $54,000]On March 30, crypto analyst Willy Woo stated that some traditional on-chain models suggest that the bottom for BTC is roughly between $46,000 and $54,000, while also indicating that a waiting period may still be necessary. The capital stored in BTC has been flowing out since November. However, it is important to note that these models have limitations. They are based on historical behavior, and Bitcoin has only experienced four bear markets in its history, all occurring against the backdrop of a long-term bull market for risk assets. If this foundation is broken, we may enter unknown territory (a deeper bear market). Ultimately, I believe that due to the potential collapse of the long-term bull market structure on a global macro level, the probability of a deeper bear market is quite high.

Willy Woo: Traditional On-Chain Models Indicate Bitcoin Bottom Between $46,000 and $54,000

On March 30, Tencent's Public Relations Director Zhang Jun announced that the WeChat Work CLI open source project has been launched on the GitHub community. It opens up seven core product capabilities including messaging, scheduling, documents, smart forms, meetings, to-do lists, and contacts, supporting mainstream AI Agents (such as ClaudeCode, Codex, WorkBuddy, QClaw, etc.). Developers can leverage these capabilities to enable AI Agents to understand and utilize WeChat Work functionalities in a more natural way, facilitating the rapid development of AI applications that are closely aligned with everyday office scenarios.

[Tencent's Zhang Jun: WeChat Work CLI Open Source Project Launched on GitHub]On March 30, Tencent's Public Relations Director Zhang Jun announced that the WeChat Work CLI open source project has been launched on the GitHub community. It opens up seven core product capabilities including messaging, scheduling, documents, smart forms, meetings, to-do lists, and contacts, supporting mainstream AI Agents (such as ClaudeCode, Codex, WorkBuddy, QClaw, etc.). Developers can leverage these capabilities to enable AI Agents to understand and utilize WeChat Work functionalities in a more natural way, facilitating the rapid development of AI applications that are closely aligned with everyday office scenarios.

Tencent's Zhang Jun: WeChat Work CLI Open Source Project Launched on GitHub

On March 30, Zhang Chi, former Chief Strategist at Guojin Securities, published a statement saying that many investors feel that the current 'US-Iran War' is similar to the 'Russia-Ukraine War.' In reality, the comparability is quite low, especially in terms of the global energy landscape and economic impact, which differ significantly. The 'Russia-Ukraine War' can at most be considered a geopolitical issue related to energy, corresponding to a localized reduction in supply. Moreover, Russia has exited the US SWIFT system but continues to sell oil, so the impact is localized and temporary. However, the core of the 'US-Iran War' is the Strait of Hormuz, where the impact on energy supply extends beyond Iran to the entire Gulf region, which holds nearly 50% of energy reserves and accounts for over one-third of energy output. I want to reiterate that the logic behind the evolution of the 'US-Iran War' is not merely about 'the exchange of national interests,' but also involves 'religious conflicts.' Therefore, there are risks of a prolonged war, a war of attrition, and escalation. The 'Strait of Hormuz' is the 'trump card' in Iran's hands, and it will not easily relinquish control—losing control of the 'Strait of Hormuz' would mean losing the war! This is the basis for my assertion that we are gradually seeing the emergence of a 'Fourth Oil Crisis' under the current 'prolonged war.' Zhang Chi defines this Fourth Oil Crisis as the 'US-Iran prolonged war' and the 'long-term blockade of the Strait of Hormuz,' which would keep oil prices high for an extended period—this assumption is crucial as it will directly determine the extent of its impact on future global inflation, economic conditions, and asset prices. Comparing the oil crises of 1973 and 1978, the structural changes in energy supply led to prolonged periods of high oil prices, lasting even 2 to 3 years; whereas the energy price increases during the '1990 Gulf War' and '2003 Iraq War' did not exceed 6 months. Clearly, whether this 'US-Iran War' can bring about a 'qualitative change' in global inflation, economic conditions, and asset prices will depend on whether the period of high oil prices is sufficiently long. My judgment is that there is indeed such a risk.

[Zhang Chi Discusses US-Iran Situation: Expectations for a Fourth Oil Crisis Are Rising]On March 30, Zhang Chi, former Chief Strategist at Guojin Securities, published a statement saying that many investors feel that the current 'US-Iran War' is similar to the 'Russia-Ukraine War.' In reality, the comparability is quite low, especially in terms of the global energy landscape and economic impact, which differ significantly. The 'Russia-Ukraine War' can at most be considered a geopolitical issue related to energy, corresponding to a localized reduction in supply. Moreover, Russia has exited the US SWIFT system but continues to sell oil, so the impact is localized and temporary. However, the core of the 'US-Iran War' is the Strait of Hormuz, where the impact on energy supply extends beyond Iran to the entire Gulf region, which holds nearly 50% of energy reserves and accounts for over one-third of energy output. I want to reiterate that the logic behind the evolution of the 'US-Iran War' is not merely about 'the exchange of national interests,' but also involves 'religious conflicts.' Therefore, there are risks of a prolonged war, a war of attrition, and escalation. The 'Strait of Hormuz' is the 'trump card' in Iran's hands, and it will not easily relinquish control—losing control of the 'Strait of Hormuz' would mean losing the war! This is the basis for my assertion that we are gradually seeing the emergence of a 'Fourth Oil Crisis' under the current 'prolonged war.' Zhang Chi defines this Fourth Oil Crisis as the 'US-Iran prolonged war' and the 'long-term blockade of the Strait of Hormuz,' which would keep oil prices high for an extended period—this assumption is crucial as it will directly determine the extent of its impact on future global inflation, economic conditions, and asset prices. Comparing the oil crises of 1973 and 1978, the structural changes in energy supply led to prolonged periods of high oil prices, lasting even 2 to 3 years; whereas the energy price increases during the '1990 Gulf War' and '2003 Iraq War' did not exceed 6 months. Clearly, whether this 'US-Iran War' can bring about a 'qualitative change' in global inflation, economic conditions, and asset prices will depend on whether the period of high oil prices is sufficiently long. My judgment is that there is indeed such a risk.

Zhang Chi Discusses US-Iran Situation: Expectations for a Fourth Oil Crisis Are Rising

On March 30, according to the latest data from institutions, the total global usage of AI large models reached 22.7 trillion tokens last week, reflecting an 11.2% month-on-month increase. Among the listed AI models, the weekly usage of Chinese AI models rose to 9.857 trillion tokens, marking a 33.94% increase from the previous week; the weekly usage of U.S. AI models was 3.007 trillion tokens, with a month-on-month growth of 1.79%. The weekly usage of Chinese AI models has increased for four consecutive weeks, consistently surpassing that of the U.S. In the past week, the top four global usage rankings were all occupied by Chinese AI models. Xiaomi's MiMo-V2-Pro maintained its top position with a weekly usage of 3.96 trillion tokens; Step 3.5 Flash (free) ranked second with a weekly usage of 1.49 trillion tokens; MiniMax M2.7 entered the rankings at third place with a weekly usage of 1.29 trillion tokens; and DeepSeek V3.2 ranked fourth with a weekly usage of 1.24 trillion tokens, reflecting an 8.7% increase. Meanwhile, MiniMax M2.5, which ranked third the previous week, fell to ninth place, nearly dropping out of the rankings, with a weekly usage of 0.909 trillion tokens, down 30% month-on-month. (Dongxin News Agency)

[China's AI Model Usage Surges Nearly 34% Month-on-Month, Outpacing the U.S. for Four Consecutive Weeks]On March 30, according to the latest data from institutions, the total global usage of AI large models reached 22.7 trillion tokens last week, reflecting an 11.2% month-on-month increase. Among the listed AI models, the weekly usage of Chinese AI models rose to 9.857 trillion tokens, marking a 33.94% increase from the previous week; the weekly usage of U.S. AI models was 3.007 trillion tokens, with a month-on-month growth of 1.79%. The weekly usage of Chinese AI models has increased for four consecutive weeks, consistently surpassing that of the U.S. In the past week, the top four global usage rankings were all occupied by Chinese AI models. Xiaomi's MiMo-V2-Pro maintained its top position with a weekly usage of 3.96 trillion tokens; Step 3.5 Flash (free) ranked second with a weekly usage of 1.49 trillion tokens; MiniMax M2.7 entered the rankings at third place with a weekly usage of 1.29 trillion tokens; and DeepSeek V3.2 ranked fourth with a weekly usage of 1.24 trillion tokens, reflecting an 8.7% increase. Meanwhile, MiniMax M2.5, which ranked third the previous week, fell to ninth place, nearly dropping out of the rankings, with a weekly usage of 0.909 trillion tokens, down 30% month-on-month. (Dongxin News Agency)

China's AI Model Usage Surges Nearly 34% Month-on-Month, Outpacing the U.S. for Four Consecutive Weeks

On March 30, Trump stated to reporters aboard Air Force One on Sunday that Iran has "met" most of the 15 ceasefire demands previously presented by the U.S. to Tehran, although it remains unclear whether substantial negotiations are actually taking place between the two sides. He said, "They have met most of our points. Why wouldn't they do it? We will put forward a few more demands." However, he declined to specify what concessions Iran had made. Publicly, Iran has clearly rejected the list of 15 ceasefire conditions and has proposed its own five preconditions, which include full control over the Strait of Hormuz. Last weekend, officials from Pakistan, Saudi Arabia, and Turkey met in an attempt to find a way out of the conflict, but there are currently no signs that the talks have made substantial progress. The Pakistani Foreign Minister stated after meeting with his counterparts from various countries, "Both Iran and the U.S. have expressed trust in Pakistan and are willing for Pakistan to host further negotiations." However, neither side has indicated readiness for direct dialogue.

[Trump: Iran Has Met Most of the 15 Ceasefire Demands]On March 30, Trump stated to reporters aboard Air Force One on Sunday that Iran has "met" most of the 15 ceasefire demands previously presented by the U.S. to Tehran, although it remains unclear whether substantial negotiations are actually taking place between the two sides. He said, "They have met most of our points. Why wouldn't they do it? We will put forward a few more demands." However, he declined to specify what concessions Iran had made. Publicly, Iran has clearly rejected the list of 15 ceasefire conditions and has proposed its own five preconditions, which include full control over the Strait of Hormuz. Last weekend, officials from Pakistan, Saudi Arabia, and Turkey met in an attempt to find a way out of the conflict, but there are currently no signs that the talks have made substantial progress. The Pakistani Foreign Minister stated after meeting with his counterparts from various countries, "Both Iran and the U.S. have expressed trust in Pakistan and are willing for Pakistan to host further negotiations." However, neither side has indicated readiness for direct dialogue.

Trump: Iran Has Met Most of the 15 Ceasefire Demands

On March 30, according to on-chain data, the Hyperliquid team wallet unstaked 333,000 HYPE tokens (approximately $13 million) 11 hours ago. HYPE is currently priced at $37.56, with a 24-hour decline of 5.14%. Previously, the official announcement stated that the 333,000 tokens from Hyperliquid Labs would be unstaked on March 29 and distributed to team members on April 6.

[Hyperliquid Labs Unstakes 333,000 HYPE Tokens for Team Distribution on April 6]On March 30, according to on-chain data, the Hyperliquid team wallet unstaked 333,000 HYPE tokens (approximately $13 million) 11 hours ago. HYPE is currently priced at $37.56, with a 24-hour decline of 5.14%. Previously, the official announcement stated that the 333,000 tokens from Hyperliquid Labs would be unstaked on March 29 and distributed to team members on April 6.

Hyperliquid Labs Unstakes 333,000 HYPE Tokens for Team Distribution on April 6

On March 30, according to Sina News, the prices of memory modules in the market have experienced a dramatic drop since last week, with the mainstream 16GB memory modules seeing a maximum daily decline of over 100 Yuan. Wholesalers generally report that this is due to large stockpilers selling off their inventory. The previous excessive price increases have led to a market sales decline of over 60% compared to before November last year, further intensifying the pressure to sell due to insufficient absorption capacity. Meanwhile, Google has released a new compression algorithm called 'TurboQuant,' which can reduce the memory usage of key-value caches during the operation of large language models by at least 60%, leading to a downward adjustment in market expectations for AI hardware demand. According to the Financial Times, as a result, the market value of U.S. memory chip stocks has evaporated by nearly $100 billion this week, with Micron Technology's stock price having dropped 15% since last Friday, resulting in a market value loss of over $70 billion.

[Dramatic Drop in Memory Prices: Mainstream 16GB Modules See Daily Decline Exceeding 100 Yuan]On March 30, according to Sina News, the prices of memory modules in the market have experienced a dramatic drop since last week, with the mainstream 16GB memory modules seeing a maximum daily decline of over 100 Yuan. Wholesalers generally report that this is due to large stockpilers selling off their inventory. The previous excessive price increases have led to a market sales decline of over 60% compared to before November last year, further intensifying the pressure to sell due to insufficient absorption capacity. Meanwhile, Google has released a new compression algorithm called 'TurboQuant,' which can reduce the memory usage of key-value caches during the operation of large language models by at least 60%, leading to a downward adjustment in market expectations for AI hardware demand. According to the Financial Times, as a result, the market value of U.S. memory chip stocks has evaporated by nearly $100 billion this week, with Micron Technology's stock price having dropped 15% since last Friday, resulting in a market value loss of over $70 billion.

Dramatic Drop in Memory Prices: Mainstream 16GB Modules See Daily Decline Exceeding 100 Yuan

Market data shows that ETH has fallen below $2000, currently priced at $1999.96, with a 24-hour decline of 0.23%. The market is experiencing significant fluctuations, so please ensure proper risk management.

[ETH Falls Below $2000]Market data shows that ETH has fallen below $2000, currently priced at $1999.96, with a 24-hour decline of 0.23%. The market is experiencing significant fluctuations, so please ensure proper risk management.

ETH Falls Below $2000

According to TechCrunch, Axelera, an AI chip company incubated by crypto mining company Bitfury, has raised $27 million in Series A round led by Innovation Industries, with participation from Imec xpand and the Federal Holding and Investment Company of Belgiu, and the financing will be used for its hardware commercialize. In addition, the Netherland Enterprise Agency awarded Axelera AI a $6.7 million loan commissioned by the Ministry of Economic Affairs and Climate Policy. Axelera is working to develop AI acceleration cards and systems for use cases like security, retail and robotics that it plans to sell through partners in the business-to-business edge computing and Internet of Things sectors.

Detail

【AI chip startup Axelera lands $27M in capital to commercialize its hardware】According to TechCrunch, Axelera, an AI chip company incubated by crypto mining company Bitfury, has raised $27 million in Series A round led by Innovation Industries, with participation from Imec xpand and the Federal Holding and Investment Company of Belgiu, and the financing will be used for its hardware commercialize. In addition, the Netherland Enterprise Agency awarded Axelera AI a $6.7 million loan commissioned by the Ministry of Economic Affairs and Climate Policy. Axelera is working to develop AI acceleration cards and systems for use cases like security, retail and robotics that it plans to sell through partners in the business-to-business edge computing and Internet of Things sectors.

AI chip startup Axelera lands $27M in capital to commercialize its hardware

Article

Santiment: Retail Sentiment Hits 'Extreme Fear,' Bitcoin Price May Rebound

TRX broke through 0.32  USDT

Fed's April Rate Hike Probability Drops to 4.1%

cfg soared to 0.1688 USDT, a new high with a 24-hour increase of 4.96%. It currently trades at 0.163 USDT on HTX.

Analysts: 'CLARITY Act' May Pose Bearish Pressure on Yield-Bearing DeFi Tokens

AI chip startup Axelera lands $27M in capital to commercialize its hardware

Isenção de responsabilidade：O conteúdo acima não representa as posições da HTX.，A HTX não faz quaisquer recomendações de negociação.。

Artigos relacionados

Todos os comentários0Mais recentePopular

Tópicos em destaque

Artigos relacionados