ASML plunges: How wide the credibility gap? An accidentally early release of a bad forecast from ASML has produced a credibility gap, with CEO Christophe Fouquet under pressure to explain why the outlook has changed so dramatically all of a sudden. It’s not clear if the
AI HARDWARE
-
ASML Shares Plunge 15% on Revised 2025 Sales Forecast
By
–
What a day… Dutch chip firm ASML released its Q3 earnings a day early – in a "technical error" – and they show it's forecasting 2025 net sales in the lower half of previous guidance… shares nosedived 15% today.
-
Databricks AWS Partnership Powers Production GenAI Applications
By
–
We’ve teamed up with @awscloud to power customized, production-quality #GenAI applications through AWS Trainium AI chips! See how Databricks + AWS allows thousands of joint customers to build, deploy and monitor high-quality, customized AI applications: https://
dbricks.co/48oIM7v -
Elon’s Starship Launch Achievement Celebrated
By
–
Thanks Elon! And congrats again on your incredible starship launch!
-
NVIDIA Blackwell Cuts Inference Costs 25x for Generative AI
By
–
Dion Harris, NVIDIA Data Center Product Marketing Head, highlights how our NVIDIA Blackwell platform provides 25x lower cost and energy consumption for generative #AI model inference than our previous generation—enabling industries to optimize #datacenter productivity. https://t.co/cDMG6woDXk
— NVIDIA (@nvidia) 14 octobre 2024Dion Harris, NVIDIA Data Center Product Marketing Head, highlights how our NVIDIA Blackwell platform provides 25x lower cost and energy consumption for generative #AI model inference than our previous generation—enabling industries to optimize #datacenter productivity.
-
Intel Xeon 6 Processors Deliver 2.6x Better Performance Per Watt
By
–
Here’s what business leaders need to know about #IntelXeon 6 processors:
— Pascal Bornet (@pascal_bornet) 14 octobre 2024
✅2.6x better performance per watt than 2nd Gen—cut energy costs
✅4.2x better rack-level performance—maximize data center space
✅Future-ready with Performance-cores coming soon for #AI & #ComputerVision… pic.twitter.com/IIvGCj6iDyHere’s what business leaders need to know about #IntelXeon 6 processors:
2.6x better performance per watt than 2nd Gen—cut energy costs
4.2x better rack-level performance—maximize data center space
Future-ready with Performance-cores coming soon for #AI & #ComputerVision -
xFasterTransformer: Optimized LLM Solution for X86 Platforms
By
–
xFasterTransformer xFasterTransformer is an exceptionally optimized solution for large language models (LLM) on the X86 platform, which is similar to FasterTransformer on the GPU platform. xFasterTransformer is able to operate in distributed mode across multiple sockets and
-
Nvidia’s Jensen Huang Hypes AI Inference Scaling Strategy
By
–
The key slide in the Nvidia roadshow that is getting people worked up. Wall Street’s Best Ideas: Nvidia’s Huang hypes the inference curve Nvidia’s Jensen Huang has been selling people on the “scaling” of inference although but you should mind the hyperbole, software’s “new
-
Tesla Optimus Robot and Robotaxi Autonomous Future
By
–
La última presentación de Tesla ha sido bastante impresionante y una ventana al futuro. Optimus, el robot de servicio que promete hacer todas las tareas del día a día. El Robotaxi, un futurista vehículo sin conductor que nos transportará por la ciudad y esta Robovan que está… pic.twitter.com/hS6bVQIMsS
— Juan Merodio (@juanmerodio) 12 octobre 2024La última presentación de Tesla ha sido bastante impresionante y una ventana al futuro. Optimus, el robot de servicio que promete hacer todas las tareas del día a día. El Robotaxi, un futurista vehículo sin conductor que nos transportará por la ciudad y esta Robovan que está
-
Building Low Level AI Infrastructure and Foundations
By
–
Very happy to be building more low level right now 🙂