
Ara-2 Discrete NPU for Gen AI - NXP Semiconductors
Oct 28, 2025 · The Ara-2 discrete neural processing units (DNPUs) enable real-time Generative AI, Large Language Models (LLMs) and Vision Language Models (VLMs) execution on AI-enabled …
Ara-2 M.2 (M-Key)模块_恩智浦半导体 | NXP 半导体
Ara-2 M.2 (M-Key)模块提供40等效TOPS (eTOPS),适用于高性能、高能效的AI推理。 该紧凑型模块旨在满足大语言模型 (LLM)、视觉语言模型 (VLM)以及基于Transformer的模型的计算需求,非常使用 …
NXP Goes After LLMs at the Edge, Aquires Kinara and Its Ara 40 TOPS ...
Feb 10, 2025 · Kinara is best known for the Ara-1 and Ara-2 neural processing units (NPUs), with the latter delivering a claimed up-to-40 tera-operations per second (TOPS) of minimum-precision …
Ara-2 M.2 (M-Key) Module | NXP Semiconductors
Designed to handle the compute demands of large language models (LLMs), Vision Language Models (VLMs) and transformer-based models, this compact module is ideal for enabling or upgrading edge …
APC880 Edge AI PC with Kinara Ara-2 and NXP i.MX8M Plus - Geniatech
Geniatech’s APC880 is a powerful edge AI computing platform, combining the NXP i.MX8M Plus SoC with the Kinara Ara-2 AI accelerator module. Designed with industrial-grade durability and flexible AI …
AI集成摄像头与嵌入式系统的Ar a-1 USB模块 Ara-1 USB Module for AI-Integrated Cameras and Embedded Systems
Ara-2 M.2 (M-Key) モジュール | NXP Semiconductors
Ara-2 M.2 (M-Key) モジュールは、最大で40実効TOPS (eTOPS) の性能を発揮し、高い電力効率でハイパフォーマンスなAI推論を実行できます。
DeepSeek R1/Llama 8B natively on NXP i.MX 8M Plus with Kinara Ara …
Mar 14, 2025 · At Embedded World 2025, NXP showcased the integration of the DeepSeek R1 8B model with its i.MX 8M Plus application processor, augmented by Kinara’s Ara-2 neural processing …
Ara-2 USB Module: eTOPS for Gen AI Workloads - NXP Semiconductors
Oct 28, 2025 · The Ara-2 USB module delivers up to 40 equivalent TOPS (eTOPS) for high-performance, power-efficient AI inference. It supports real-time AI workloads, including Generative …
Ara2 Runs Generative AI On Edge - TechInsights
Dec 4, 2025 · The new Ara2 neural-processor unit (NPU) from Kinara, which means edge in Hindi, provides an 8× increase in performance over the first-generation chip and can run large language …