We are happy to announce that we have brought up support for Llama-3.1-70B inference on Tenstorrent’s 8-chip systems, the TT-QuietBox and the TT-LoudBox. The source code for Llama-3.1-70B and other models that are supported is on our GitHub —>
AI HARDWARE
-
AI Investment Focus: Nvidia Clusters Over Current DoE Infrastructure
By
–
I wish they would spend their AI money on buying megaclusters of Nvidia instead. Even the clusters at DoE labs seem small compared to what is being built in the commercial world.
-
GPU Configuration and Memory Optimization for Model Fine-tuning
By
–
The process of fine-tuning and evaluation involves several technical challenges: • Configuring GPU computing environments
• Optimising memory usage and finding the optimal batch size
• Setting up model configurations for specific requirements -
Physical Embodied AI Taking Off Insights
By
–
Thanks to @CarinaNamih @fankhauser @ahtih and @salar for their insights on why physical/embodied AI is taking off. Read w/
@CristinaCriddle -
Toronto researchers develop single-photon camera for advanced imaging
By
–
Researchers at the University of Toronto have developed a single-photon camera (SPAD) that captures and reconstructs light at any timescale. Explore the research [
https://
spr.ly/6018gqs3y -
Six Petabytes of Data Per Day Driving AI Infrastructure
By
–
6… petabytes… of data… per day… h/t @_philschmid
-
Metis AIPU Edge AI Vision Technology Showcased at Summit
By
–
Check out Bram Verhoef showcasing our edge AI and vision technology from the 2024 Embedded Vision Summit – @edgeaivision ! See the Metis AIPU's 200+ TOPs and high-performance computer vision solutions. Watch now: https://
youtube.com/watch?v=mNkL_J
ADTm4
…
#AI #ComputerVision #Innovation -
Jensen Huang Named Top 35 HPC Legend by HPCWire
By
–
We are honored to share that our CEO, Jensen Huang, has been named a top 35 #HPC legend by @HPCWire for his visionary leadership and groundbreaking advancements in accelerated computing. https://
nvda.ws/4fgsbFP -
LLMs Reaching LHC-Level Complexity in Infrastructure and Development
By
–
LLMs as an artifact are trending to the complexity of something like the LHC. This is clear when you look at the datacenter computronium build out but it's a lot more than that – a large chunk is digital and much harder to see/appreciate, it's just a bunch of people on a laptop.
-
NVIDIA Accelerated Computing Delivers Energy Efficiency and Cost Savings
By
–
Accelerated computing and #AI are delivering #energyefficiency while reducing costs for many industries. Explore how NVIDIA technologies are tackling power demands from data centers and why accelerated computing is sustainable computing on our blog: https://t.co/raVmMG3Rcn pic.twitter.com/RSUmZduUi6
— NVIDIA (@nvidia) 24 juillet 2024Accelerated computing and #AI are delivering #energyefficiency while reducing costs for many industries. Explore how NVIDIA technologies are tackling power demands from data centers and why accelerated computing is sustainable computing on our blog: https://
nvda.ws/3Wwjujc