LPU

¿Muy poco y demasiado tarde? Bertha LPU se une a la LPU ultrarrápida de Groq a medida que crece el desafío al gigante de GPU de Nvidia

by

[ad_1] Una startup de inteligencia artificial en Corea del Sur Hiper Axel Se asoció con la empresa de SoC basada en plataforma y el diseñador de ASIC SEMIFIVE en enero de 2024 para desarrollar Bertha LPU. Diseñado para la inferencia LLM, Bertha ofrece “bajo costo, baja latencia y características específicas de dominio”, con el objetivo … Read more

This startup wants to take on Nvidia with a server-on-a-chip to eliminate what it calls an already flawed system — faster GPU, CPU, LPU, TPU or NIC will not deliver the leap that many firms are aiming for

by

[ad_1] According to Israeli startup NeuReality, many AI possibilities aren’t fully realized due to the cost and complexity of building and scaling AI systems.  Current solutions are not optimized for inference and rely on general-purpose CPUs, which were not designed for AI. Moreover, CPU-centric architectures necessitate multiple hardware components, resulting in underutilized Deep Learning Accelerators … Read more

Groq LPU (Language Processing Unit) performance tested – capable of 500 tokens per second

by

A new player has entered the field of artificial intelligence in the form of the Groq LPU (Language Processing Unit). Groq has the remarkable ability to process over 500 tokens per second using the Llama 7B model.  The Groq Language Processing Unit (LPU), is powered by a chip that’s been meticulously crafted to perform swift inference … Read more