英伟达市值重返5万亿,比特币现货ETF连续9日净流入

NVIDIA Market Cap Returns to $5 Trillion, Bitcoin Spot ETF Sees Net Inflows for 9 Consecutive Days

BroadChainBroadChain04/25/2026
This content has been translated by AI
Summary

NVIDIA's market cap returns to $5 trillion, while U.S. spot Bitcoin ETFs have recorded net inflows f

BroadChain learned that at 19:02 on April 25, Nvidia's stock price rose on the day of DeepSeek V4's release, pushing its market cap back above $5 trillion. Market optimism about AI computing demand continues, despite Chinese large model DeepSeek V4's performance being comparable to Gemini 3 Flash at just one-fifth the cost. The community is hotly debating whether open-source models can shake Nvidia's position and how the business landscape will be reshaped.

The U.S. Treasury froze $344 million in USDT linked to Iran, with Tether cooperating in the execution. However, analysts point out that if Bitcoin were used, the U.S. could not freeze any transaction. This comparison highlights the fundamental difference between centralized stablecoins and decentralized cryptocurrencies, with the community intensely discussing USDT's role as a dollar tool versus Bitcoin's censorship resistance.

U.S. spot Bitcoin ETFs recorded a net inflow of $14.45 million on April 24, extending a nine-day winning streak. On the same day, Ethereum ETFs saw a net inflow of $23.38 million. Additionally, a trader opened a $50.6 million BTC short position ahead of Trump's crypto policy speech, betting on a "sell the news" scenario.

Approximately 35 million XRP were withdrawn within 24 hours, with large-scale withdrawals typically seen as a bullish signal due to reduced exchange selling pressure. U.S. crypto company Bitmine purchased 10,000 ETH for $23.9 million, now holding 4.97 million ETH, making it the largest public ETH holder globally.

The Philadelphia Semiconductor Index set a record with 18 consecutive gains, with the market focusing on AI bottleneck areas such as CPUs and optical communications. The surge in AI inference demand has brought CPUs back to a core position, benefiting related A-share companies. Google DeepMind released the open-source Gemma 4 model, claiming leading "byte-level performance," competing with Meta Llama and DeepSeek.