I am 98.2% sure this will be us.......DYOR!
View attachment 34838
"Arun Iyengar, CEO of Toronto-based tech firm Untethered AI, explains that traditional AI methods are a huge drain on the world’s energy resources.
“Just looking at January, if you look at the number of people that used ChatGPT and the amount of energy it took to serve them, it would be the same amount of energy to fully serve a town of 175,000 people,” he says.
That’s because it’s using methods established as far back as the 1940s by John von Neumann, widely considered the father of modern AI technology, which has powered all computer chips to date.
“Anytime you move data you burn energy. So the more distance you move data, the more power you’re burning to make that happen. And that’s why today’s implementations end up consuming so much power.”
Iyengar’s company, a semiconductor startup, has created a first-of-its-kind chip that aims to dramatically reduce AI’s energy use.
“We are basically putting the processing and the memory right next to each other. So the data movement is now contained in a very,
very short distance … that reduces the power or the energy required to do Artificial Intelligence by a factor of six to 10 times,” he says"
Ella and I are 99.9° % sure that Untether have their own in-house design:
US2021091794A1 COMPUTATIONAL MEMORY WITH ZERO DISABLE AND ERROR DETECTION
A processing element includes an input zero detector to detect whether the input from the neighbor processing element contains a zero. When the input from the neighbor processing element contains the zero, a zero disable circuit controls the input from the neighbor processing element and respective data of the memory to both appear as unchanged to the arithmetic logic unit for the operation. A controller of an array of processing elements adds a row of error-checking values to a matrix of coefficients, each error-checking value of the row of error-checking values being a negative sum of a respective column of the matrix of coefficients. The controller controls a processing element to perform an operation with the matrix of coefficients and an input vector to accumulate a result vector. Owing to the error-checking values, when a sum of elements of the result vector is non-zero, an error is detected.
Technology — Untether AI
https://www.untether.ai/technology
runAI200™ Represents a new breed of computing
At Untether AI, we re-wrote the rules for compute architectures. Designed from the ground up for AI inference workloads, the runAI200® architecture provides best-in-class performance for running neural networks; CNNs, RNNs, TCNs, Attention, Transformer, Unets, and DLRM.
At the heart of the unique at-memory compute architecture is a memory bank: 385KBs of SRAM with a 2D array of 512 processing elements. With 511 banks per chip, each device offers 200MB of memory, enough to run many networks in a single chip. And with the multi-chip partitioning capability of the imAIgine® Software Development Kit, larger networks can be split apart to run on multiple devices, or even across multiple tsunAImi® accelerator cards.