BTC $62,675.28 -1.40%
ETH $1,872.62 -0.72%
BNB $604.42 -0.74%
XRP $1.00 -0.39%
SOL $75.38 -0.76%
TRX $0.3330 -0.26%
DOGE $0.0691 -1.31%
ADA $0.1792 -2.08%
BCH $203.03 -4.93%
LINK $8.85 +0.90%
HYPE $56.10 -2.52%
AAVE $85.60 -3.44%
SUI $0.6758 -1.83%
XLM $0.1591 -0.53%
ZEC $483.57 -1.16%
BTC $62,675.28 -1.40%
ETH $1,872.62 -0.72%
BNB $604.42 -0.74%
XRP $1.00 -0.39%
SOL $75.38 -0.76%
TRX $0.3330 -0.26%
DOGE $0.0691 -1.31%
ADA $0.1792 -2.08%
BCH $203.03 -4.93%
LINK $8.85 +0.90%
HYPE $56.10 -2.52%
AAVE $85.60 -3.44%
SUI $0.6758 -1.83%
XLM $0.1591 -0.53%
ZEC $483.57 -1.16%

evaluation

All
Article
Flash

hot_img Samsung Electro-Mechanics' customer evaluation for glass substrates encounters obstacles, delaying equipment investment schedule

According to The Elec, Samsung Electro-Mechanics' investment schedule for semiconductor glass substrates has been delayed due to bottlenecks in the reliability assessment of customer prototypes. As a result, the construction schedule for the production line of GlaSSEM, the glass core joint venture between Samsung Electro-Mechanics and Dongyu Fine Chem, has also been postponed, and the timeline for ordering related equipment has become uncertain.The report states that GlaSSEM has failed to communicate the equipment order schedule to its partner manufacturers. Previous orders have been postponed from December last year to March and June this year, and this has happened more than three times. After June, no further schedule guidance has been provided. Industry insiders believe that the main reason is that global communication semiconductor customers have not passed the reliability assessment certification for glass substrate prototypes. Samsung Electro-Mechanics is still producing prototypes at its Sejong plant, and most of the suppliers for key processes required for building the mass production line, such as deposition, etching, laser drilling, and inspection, have already been selected.Since the prototypes need to be remade and resubmitted, the originally planned investment schedule will inevitably be further delayed. Samsung Electro-Mechanics had set a target for mass production in the second half of 2027 when establishing GlaSSEM earlier last month, but during last month's financial report meeting, this timeline was adjusted to 2028, and now the mass production time is expected to be further postponed.

Data: 20% of active BTC investors are currently in a state of unrealized losses, and the market has entered a mild devaluation range

Cryptocurrency analyst Darkfost published an on-chain indicator interpretation on the X platform, analyzing the current profit and loss structure of Bitcoin positions in the market through the core economic indicator TMM (True Market Mean) and the AVIV ratio. The TMM indicator excludes long-term untransferred and partially permanently lost dormant Bitcoins, only accounting for the average holding cost of actively traded chips in the market, with the current value around $76,700. This level has become a significant price resistance, as during the May market, when the price reached this point, many investors exited to break even.The accompanying AVIV (Active Value / Investor Value) ratio currently hovers around 0.8, within the valuation discount range, indicating that the average floating loss for all active BTC investors is 20%. Historically, at the bottom of bear markets, this indicator has dropped to 0.5 to 0.6, corresponding to investor losses of 40%-50%. The current loss magnitude has not yet reached extreme bear market levels.The analyst also pointed out that even with a large influx of institutional funds and Bitcoin ETFs injecting massive liquidity in this cycle, Bitcoin still follows its own cyclical patterns. In the short term, there is no need to wait for the indicators to drop to historically extreme low levels before a rebound occurs, but it is necessary to acknowledge the current pressured environment of widespread losses among market participants.

DGrid AI released the latest research paper PoQ-Judge, completing the closed loop of decentralized LLM quality assessment with a multi-architecture evaluation framework

The decentralized AI infrastructure network DGrid AI today released its latest research paper "PoQ-Judge," proposing a multi-architecture quality assessment framework that does not require reference answers. This means that in real deployment environments, there are often no standard answers for comparison, yet the protocol can still reliably score the quality of model responses and allocate incentives accordingly. This is a key piece that has long been missing in DGrid's decentralized LLM inference quality assessment system.PoQ (Proof of Quality) is a consensus mechanism independently developed by DGrid, designed to prevent model providers from deploying low-quality models, fabricating data, or hiding computational costs at the protocol level, thereby ensuring service quality and pricing transparency. The DGrid team has been continuously working on PoQ and has published four research papers to date. The newly released PoQ-Judge has trained three assessment models covering different quality and cost scenarios, achieving a correlation of up to 0.747 with human scoring on the retention test set, significantly outperforming all previous reference answer-based evaluators, while reducing assessment costs by over 72% through cascading evaluation and online weight calibration.With the implementation of PoQ-Judge, the entire process from quality assessment → scoring → incentive allocation has completely eliminated reliance on reference answers, thus establishing a closed loop for the quality of decentralized LLM inference.DGrid AI is a decentralized AI intelligent network dedicated to building an open, transparent, and community-driven AI infrastructure. Focusing on model invocation and application experience, DGrid has launched several core products: the AI Gateway that aggregates mainstream large models globally, the one-click deployment platform for AI agents DClaw, the anonymous model competition platform AI Arena, and the intelligent model recommendation assistant Dori, providing one-stop services for developers and users. It is reported that DGrid AI's revenue has surpassed 20 million dollars in six months.
app_icon
ChainCatcher Building the Web3 world with innovations.