Open Account
Demo Account
About Us
Real-time Quotes & News
Market Analysis
Economic Calendar
Daily Market Analysis
Trading Platform
Platform Overview
How To Use
Terms
All Terms
Deposit & Withdrawal
Promotion
FAQ
Contact
繁
简
EN
User Login
Open Account
Demo Account
繁
简
EN
User Login
Open Account
Demo Account
About Us
About Aspire
Features of Aspire
Real-time Quotes & News
Real-time Quotes
Real-time News
Market Analysis
Economic Calendar
Market Analysis
Trading Platform
Meta Trader 5
Platform Features
Terms
All Terms
Deposit & Withdrawal
Promotion
FAQ
Contact
About Us
Terms
Metals Market
Trading Platform
Market Analysis
Promotion
FAQ
Contact
繁
简
EN
EUROSTOXX 50 futures up 0.45%; DAX futures up 0.50%; CAC 40 futures up 0.39%; FTSE futures up 0.33%.
2026-10-12
EUROSTOXX 50 futures up 0.45%; DAX futures up 0.50%; CAC 40 futures up 0.39%; FTSE futures up 0.33%.
Back
Other News
2026-10-12
According to data from South Korean customs, exports from October 1 to 10 increased by 102.0% year-on-year, with semiconductor exports increasing by 170.9% year-on-year.
According to data from South Korean customs, exports from October 1 to 10 increased by 102.0% year-on-year, with semiconductor exports increasing by 170.9% year-on-year.
2026-10-12
In his Interconnects newsletter, AI researcher Nathan Lambert wrote that many aspects of the AI training and inference stack are highly verifiable. Training metrics, such as tokens per GPU per second, essentially represent training speed. Inference m
In his Interconnects newsletter, AI researcher Nathan Lambert wrote that many aspects of the AI training and inference stack are highly verifiable. Training metrics, such as tokens per GPU per second, essentially represent training speed. Inference metrics, such as tokens per prompt and FLOPs per token, ultimately reflect the cost per answer. Both are highly optimizable, especially with mature sub-problems and architectural tradeoffs to leverage. I anticipate that within a few years, AI agents will help optimize the entire process end-to-end, bringing our inference capabilities close to the underlying computing power limits of accelerators like GPUs. In recent years, companies have already seen significant gains in inference efficiency—for example, after releasing a model at a certain price, they can save an additional 10–30% in service costs. The entire stack will continue to compound, and I expect the effective cost of model intelligence to decrease almost exponentially in the coming years (possibly faster than recent trends). These low-hanging fruits of efficiency can be harvested within a few years. In the longer term, the co-design of accelerators and models has an even longer timeline, potentially bringing several orders of magnitude more efficiency improvements on flexible GPU platforms. During this period of rapid efficiency improvements, the flexibility of GPUs will be beneficial because the architectural exploration space is crucial and a major lever for automated research. In my view, it is reasonable to predict that pre-training research (at least in terms of architecture and data selection for current models) will be automated within 2–3 years.
Chat with us
, powered by
LiveChat