#JUSTADDED to our YouTube channel! Join our #VPofProduct, @beach66
, to discover how our #speedAI family of AI #accelerator chips can deliver over 2 PFlops of #AI #inference performance with a new #FP8 datatype at 30 TOPS/W energy efficiency. https://
bit.ly/3F1rHTu
@untetherai
-
SpeedAI Accelerator Chips Deliver 2 PFlops AI Inference Performance
By
–
-
FP8 Precision Breakthrough: Advanced AI Inference Acceleration
By
–
Industry's only advanced way to get 8-bit precision with the highest #accuracy and highest #throughput. Download our FP8 whitepaper on http://
untether.ai to learn more. #AI #energyefficiency #inference #acceleration #NeuralNetworks -
INT8 Transformers Optimization for Inference Acceleration
By
–
#FBF to our @NeurIPSConf 2022 paper presentation on #INT8 #Transformers for #InferenceAcceleration. View our presentation* at https://
bit.ly/3DzONzG and download the paper at http://
untether.ai/resources. (*Playback requires Google Chrome browser) #AI #machinelearning #ENLSPII -
Top 20 Canada Tech Provider Accelerates AI Inference
By
–
Honored to be chosen as one of @evbizmag
's "Top 20 Canada Technology Solution Providers 2022"! We're committed to making a difference in #tech by accelerating #AI #inference for diverse #neuralnetworks using our #energycentric computing architecture. https://
bit.ly/3wIBmtx -
Energy-Centric AI Acceleration with Chiplet-Based Design
By
–
Join us this evening at the #ChipletSummit for a featured paper presentation by our #VPofProduct, @beach66
! "Energy-Centric #AI #Acceleration with Chiplet-Based Design" (
https://
bit.ly/3R6dXvi) A-103: Design-1 Session 4:30 – 5:30 PM (Pacific Standard Time) -
Energy-Centric AI Acceleration with Chiplet-Based Design Summit
By
–
Join us at the #ChipletSummit on 1/25 @ 4:30PM for the Design-1 Session! @beach66
, #VPofProduct at @UntetherAI
, will present his paper, "Energy-Centric #AI Acceleration with Chiplet-Based Design." Learn more about the event and register now at https://
bit.ly/3iStbHQ! -
Untether AI SDK enables bare-metal neural network HPC programming
By
–
@UntetherAI
's #newest version of imAIgine® #SDK now allows bare-metal programming for customers in fast-moving #neuralnetwork applications or #HPC. Learn more via @sallywf
's latest article in @eetimes and our press release on http://
untether.ai. https://
bit.ly/3WmvgcC -
Energy-Centric AI: Efficient ML Models Without Precision Loss
By
–
How can #ML/#DL models run efficiently without precision? In the latest @SemiEngineering article, @UntetherAI
's @beach66 & Martin Snelgrove introduce the Energy-Centric AI computing architecture: a solution built for energy-efficient #AI inference. -
AI Inference Hardware Optimization for Machine Learning Workloads
By
–
Thank you for your continued support as we work to help companies run AI inference workloads faster, cooler, and more cost-effectively using #atmemorycomputation. We look forward to the innovations and advancements that 2023 will hold! #ML #deeplearning #neuralnetworks #AI
-
UntetherAI Selected for Vector Institute AI20 Program
By
–
Thank you, @VectorInst
, for selecting us as a member of the Vector #AI20! As we look forward to 2023, we will continue to use the disruptive thinking that @UntetherAI was founded on to engineer solutions that better the Canadian and global communities. https://
bit.ly/3UV7OTe