Groq offers LPU
Product
Groq LPU is a custom inference chip with on-chip SRAM and deterministic scheduling delivering low-latency AI inference for LLMs.
- Currently Offered
Groq LPU is a custom inference chip with on-chip SRAM and deterministic scheduling delivering low-latency AI inference for LLMs.
The Groq LPU (Language Processing Unit) is a tensor streaming processor that combines compute, SRAM, and networking on a single die. It uses a software-scheduled, deterministic execution model designed for predictable latency and high throughput on generative AI inference workloads.
In December 2025, Nvidia entered a non-exclusive licensing agreement for Groq's inference technology, validating the LPU architecture. Groq continues to operate independently and in June 2026 raised $650M to scale its inference cloud business.
Cerebras Wafer-Scale Engine is a family of wafer-scale AI processors.
Analog compute-in-memory processor delivering 25 TOPS for edge AI inference.
Analog in-memory inference platform collocating deep sub-threshold compute inside multi-level non-volatile memory.
Maverick-2 is NextSilicon's runtime-reconfigurable dataflow accelerator for high-performance computing and AI workloads.
PowerEdge XE9680 is Dell's 6U GPU-dense server for AI/ML supporting up to eight NVIDIA, AMD, or Intel accelerators.
Trainium is AWS's purpose-built AI accelerator chip family for training and inference of large generative models.
Licensable AI processor IP for on-device inference across edge silicon designs.
Cerebras Wafer-Scale Engine is a family of wafer-scale AI processors.
Analog compute-in-memory processor delivering 25 TOPS for edge AI inference.
Analog in-memory inference platform collocating deep sub-threshold compute inside multi-level non-volatile memory.
Maverick-2 is NextSilicon's runtime-reconfigurable dataflow accelerator for high-performance computing and AI workloads.
Custom AI model builder for enterprise infrastructure.
NVIDIA Blackwell is a data-center GPU architecture for generative AI and accelerated computing.
NVIDIA Hopper is a data-center GPU architecture for AI and high-performance computing workloads.
NVIDIA B200 GPU for high-performance AI training and inference.
PowerEdge XE9680 is Dell's 6U GPU-dense server for AI/ML supporting up to eight NVIDIA, AMD, or Intel accelerators.
TPU is Google's custom AI accelerator ASIC family delivered through Google Cloud TPU Pods for ML training and inference.
Trainium is AWS's purpose-built AI accelerator chip family for training and inference of large generative models.
Intel Gaudi is a data-center AI accelerator product family.
