@SandiaLabs and Cerebras Systems have unveiled a cluster composed of four Cerebras CS-3 systems to be used as a Sandia testbed. This third-generation wafer scale engine architecture (WSE-3) will expand the capabilities of the NNSA Tri-labs and will allow bleeding edge
AI HARDWARE
-
SambaNova Cloud Platform Speeds Validated by AI Community
By
–
@ArtificialAnlys just put our speeds to the test and we're loving the results! Thank you, to the real MVP of #AI validation! Try it for yourself at https://
cloud.sambanova.ai @AIatMeta -
CoreWeave Reaches $23 Billion Valuation in Secondary Round
By
–
CoreWeave is officially worth $23 billion now as Cisco and Pure Storage participate in a secondary
-
AMD Peer-to-Peer Bandwidth Performance Impact on Distributed Inference
By
–
If you're evaluating which accelerator to choose for your future workloads beware that AMD's peer-to-peer bandwidth is 7x slower than all-to-all bandwidth. As long as you use full nodes or single gpus, you have nothing to worry about. But if you have to deploy TP=2, TP=4, or
-
SambaNova Cloud Accelerates Llama 3.2 AI Inference Performance
By
–
Reach for the stars with SambaNova Cloud!
— SambaNova (@SambaNovaAI) 13 novembre 2024
Unlock fast #AI inference on @AIatMeta's Llama 3.2 1B & 3B with unmatched performance — all running at full precision.
Start building today ⤵️Reach for the stars with SambaNova Cloud! Unlock fast #AI inference on @AIatMeta
's Llama 3.2 1B & 3B with unmatched performance — all running at full precision. Start building today -
China’s Electric Humanoid Robot Tiangong Goes Open Source
By
–
World’s first electric humanoid robot goes open-source to speed up integration And once again its China. I cannot but see them as way ahead when it comes to robotics. "The world’s first full-size purely-electric running robot, Tiangong from China, has gone open source. The
-
Elon Musk’s Supercomputer Colossus Disrupts AI Industry Competition
By
–
How Elon Musk’s Supercomputer Freaked Out AI Rivals As promised, after subscribing to the Information, ill keep you updated. Elon Musk has shaken the AI industry by rapidly building a supercomputer, “Colossus,” for his xAI venture, prompting rivals like OpenAI and Microsoft to
-
LLM Inference Costs Declining Through Hardware and Optimization Advances
By
–
The falling cost of LLM inference The costs of running LLMs are high, especially during inference. But an a16z’s analysis shows that “LLMflation” (the rising cost of AI use) could be curbed as advancements make AI more efficient. With innovations in hardware, optimization
-
US Nuclear Power Expansion Essential for AGI Development
By
–
US Unveils Plan to Triple Nuclear Power by 2050 as Demand Soars. This is good and necessary for AGI! As I have already written several times in various analyses, compute and energy are the biggest problems on the way to AGI / ASI. https://
forwardfuture.ai/p/scale-is-all
-you-need-part-3-2
…
On the one hand, because -
Molecular Dynamics Timescale Barrier Broken with Wafer-Scale System
By
–
We are excited to be finalists for the 2024 Gordon Bell Prize with our partners @SandiaLabs
, @Livermore_Lab
, and @LosAlamosNatLab
. This year we are proud to present our collaborative work: Breaking the Molecular Dynamics Timescale Barrier Using a Wafer-Scale System. Winners