The "Big Short" Michael Burry’s latest portfolio adjustment: he increased his holdings of JD.com’s 2029 call options and BYD’s common stock.

According to on-chain monitoring, the US government has just transferred funds seized from the Bitfinex hack, with roughly 264.863 BTC (valued at approximately $22.87 million) moved to a new address.

According to on-chain monitoring, the U.S. government has just transferred funds seized from the Bitfinex hack, with approximately 264.863 BTC (worth around $22.87 million) moved to a new address.

Winklevoss has filed a listing application for a ZEC spot ETF with the U.S. SEC, under the ticker symbol WINK.

Oct 6 Update: #Bitcoin ETFs: 1D NetFlow: -1,059 $BTC(-$91.72M)?? 7D NetFlow: +1,074 $BTC(+$93M)?? #Ethereum ETFs: 1D NetFlow: -21,432 $ETH(-$58.29M)?? 7D NetFlow: -79,193 $ETH(-$215.37M)??

Nasdaq-listed DeFi Development Corp. has authorized the launch of a CHAD preferred stock repurchase program, which allows for the repurchase of all outstanding CHAD preferred shares, including any future shares the company may issue. The program enables the company to conduct repurchases flexibly when CHAD’s share price falls below its $10 par value per share. The Solana treasury-focused company stated it has no immediate plans to repurchase CHAD, noting it first wants the security to build market liquidity and stabilize at or near its $10 par value. This indefinite authorization adds a potential capital allocation tool to DeFi Development (DFDV)’s existing $300 million CHAD ATM issuance program. The company previously said it plans to raise funds via this program to purchase additional SOL, and paid CHAD’s first dividend on October 1.

Beating AI Express News: AI lab Banbury Road has released Kardashev-0.7, a system composed of 32 distinct models. The team trained these models together via reinforcement learning, without predefining each model’s specific tasks, enabling them to develop unique specialties and complementary capabilities during training. This methodology is called RL for Population Scaling (RLPS). Prior RLPS experiments only reached up to 16 models. In the 8-model trial, the jointly trained model population scored 81.70%, outperforming 71.04% for 8 independently trained models and 72.65% for the same model sampled 8 times. In the 16-model experiment, 22 questions were answered correctly by only one model each, verifying that different models did learn distinct capabilities. Kardashev-0.7 expands this to 32 models. Banbury Road claims it delivers state-of-the-art performance at 0.007 to 0.02x inference cost and 0.03x memory usage. Its API beta is currently on a waitlist. However, the population scores from the earlier 8 and 16-model trials were derived post-hoc by picking the best result from multiple model outputs using standard answers. Since real-world applications lack standard answers, Banbury Road has not yet disclosed a full solution for the system to automatically select correct responses.