SushiRebalance

$NVDA One Report Could Move the AI Sector
NVIDIA’s earnings can influence far more than its own stock. A strong report could lift semiconductor and AI-related names, while weak guidance could pressure the entire technology trade. This is a major market event.
#NVIDIAEarningsWeek

$BTR /USDT Perp – "Vertical Pump – Watch"
Trading Plan Watch (High Risk)
Entry: N/A
SL: N/A
TP1: N/A
TP2: N/A
BTR is up +41.93% at 0.03192 after a 34% surge in 15 minutes. It is completely disconnected from the EMAs. MACD shows a spike, but this is extremely volatile. Avoid FOMO. Wait for the pullback to the EMA10 (0.02495) for a safer entry if the uptrend holds. This is a high-risk pump-and-dump scenario.
#KIMIPreIPOsNowOpen

Crypto界 News, according to Iran’s Mehr News Agency: Iran launched missiles and drones at Kuwait and Bahrain.

CoinWorld news, SWIFT announced the launch of a blockchain-based system to enable 24/7 cross-border payments, and 17 global banks are preparing to pilot real-time transactions using tokenized deposits.

$BTC Building a community together🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧🧧towards financial freedom!

CoinWorld News: The Romanian Ministry of National Defense reported that fragments of a drone that had been safely detonated were found in Tulcea County near the border with Ukraine. Preliminary investigation results indicate that the fragments are remnants of weapons used by Russia in its attack on Ukraine in April this year.

CoinWorld News: According to Tenarmoralert monitoring, the AIDC token on the BSC chain suffered a suspicious attack, resulting in losses of approximately 121,100, i.e., 12.11 ten-thousand RMB.

This article argues that in crypto industry hiring, verification should be prioritized, relegating trust signals like resumes to a secondary role, and directly assessing capability through verifiable outputs. Evidence includes public code, testnet deployments, on-chain activities, hackathons, and DAO contributions. On-chain and open-source data should be integrated into the hiring process, placing candidates in real-world scenarios such as core protocol libraries and EIP discussions. Credentials and experience still hold value, but they must be combined with actual output.

CoinWorld News, Wu said that on-chain detective Zachxbt tracked and found that the stolen funds from Humanity Protocol (loss of $32 million) and Kelp DAO (loss of $292 million) have been mixed, indicating that the hackers behind the two incidents may overlap. Zachxbt stated that based on this new evidence, the suspicion of an insider job in the Humanity Protocol theft can basically be ruled out.

Coin World News, Ukraine's national oil company: Russia's attack severely damages its natural gas production and storage facilities.

CryptoWorld News reports that, according to Whale Alert monitoring, 2,317 BTC (approximately $144.68 million) were transferred from an unknown wallet to a Coinbase institutional account.

CoinJie.com news reports that SUI processed stablecoin transfers totaling more than $65 billion in less than a week.

According to The Bangkok Post, Thai police raided a luxury residence in Salaphi District, Chiang Mai Province, and arrested 10 Chinese suspects (9 men and 1 woman).
Seized on site were 67 mobile phones, 7 iPads, and 13 laptops.
The police stated that the gang impersonated insurance company and courier company employees, luring Chinese victims to install software that could access personal information and carry out scams.
The proceeds from the scams were transferred and settled through cryptocurrency channels.
Some suspects entered Thailand on tourist visas and rented the involved property for approximately 80k Thai Baht per month.

JetBrains open-source Mellum-2, a 12B parameter code-mixed expert model. To control inference costs, only about 2.5 billion parameters are activated per token, with weights already released on Hugging Face, Apache 2.0. A new multi-token prediction (MTP) module has been added, where the MTP head acts as a draft model to accelerate sampling during inference. Provides three versions: basic, dialogue, and thinking; the thinking version can display explicit reasoning chains before output. Benchmarks: Humaneval 41.46%, MMLU 70.87%.