Mythic: How an Analog Processor Could Revolutionize Edge AI Alberto Romero, LLM Analyst Karl Freund, Founder and Principal Analyst Cambrian-AI Research.
Mythic’s ACE meets the requisites of edge AI inference
Mythic’s tech promises high performance at very low power, ultra-low latency, low cost, and small form factor. The basic element is their Analog Matrix Processor which features an array of tiles, each containing the ACE complemented by digital elements: SRAM, a vector SIMD unit, a NoC router, and a 32-bit RISC-V nano processor. The innovative design of the ACE eliminates the need for DDR DRAM, reducing latency, cost, and power consumption.
The company’s full-stack solution leverages the potential of analog processors while maintaining relevant features of the digital world. It makes the M1076 AMP a great option to handle AI workloads for inference at the edge faster and more efficiently — the company claims it provides the “best-in-class TOPS/W” — than its fully-digital counterparts.
So far, it seems Mythic has transformed an innovative idea into promising tech to compete for edge inference AI. Now, let’s see the numbers. The company claims the M1076 AMP performs at up to 25 TOPS running at around 3W. Compared to similar digital hardware, that’s a reduction in power consumption of up to 10x. And it can store up to 80M weights on-chip. The MP10304 Quad-AMP PCIe card can deliver up to 100 TOPS at 25W and store 320M weights.
As Mythic continues to refine its hardware and software, we will look forward to seeing benchmarks that can demonstrate the platform’s capabilities and power efficiency. But we have seen enough already to be excited by the potential of this unique approach.
United States Latest News, United States Headlines
Similar News:You can also read news stories similar to this one that we have collected from other news sources.
AI Ethics Confronting Whether Irate Humans That Violently Smash Or Mistreat AI Is Alarmingly Immoral, Such As Those Angered Folks That Lash Out At Fully Autonomous AI SystemsConsider the AI ethics of humans that mistreat AI, including people that lash out at AI-based self-driving cars.
Read more »
NeMo Megatron Reinforces NVIDIA AI Leadership In Large Language ModelsNeMo Megatron Reinforces NVIDIA AI Leadership in Large Language Models Alberto Romero, LLM Analyst Karl Freund, Founder and Principal Analyst Cambrian-AI Research.
Read more »
Leicester student behind Covid AI tool completes PhD earlySiyuan Lu, from China, developed the tool together with his PhD supervisor, Prof Yudong Zhang.
Read more »
Researchers at the Columbia University may have developed an AI-driven algorithm that could prove revolutionarResearchers at the Columbia University may have developed an AI-driven algorithm that could prove revolutionary for the future of physics.
Read more »
AI: The Somnium Files - Nirvana Initiative Review: A Detective MasterpieceComingSoon's AI: The Somnium Files - Nirvana Initiative review takes a look at the Kotaro Uchikoshi-penned game and why it's his best story yet.
Read more »