On-device AI– tag –
-
Qualcomm Unveils Next-Gen Hexagon NPU with 50% Memory Increase for Accelerated Agentic AI, Potentially in Galaxy S27 Ultra
SammyGuru USA Overview Qualcomm announced its next-generation Hexagon NPU for upcoming Snapdragon platforms, significantly enhancing the speed and capability of agentic AI. This new AI hardware, expected to be featured in devices like th... -
Samsung Research Unveils AnySimLite: Sub-700KB, Sub-30ms On-Device AI Matching 7B-Parameter LLMs for Speech Classification
Samsung Research South Korea Overview Samsung Research has introduced AnySimLite, a lightweight few-shot similarity encoder capable of delivering state-of-the-art performance for multiple on-device speech-adjacent classification tasks. O... -
Microsoft Copilot+ PC Mandates Minimum 40 TOPS NPU, Redefining AI PC to Accelerate On-Device AI Processing
Layer3Labs USA Overview The definition of an AI PC is evolving, with the integration of Neural Processing Units (NPUs) to execute AI models directly on the device, rather than solely in the cloud, becoming a central requirement. Microsof... -
Qualcomm Unveils Snapdragon X7 Gen 2: Doubles NPU AI Processing Power for Enhanced On-Device AI in Next-Gen AI PCs and Premium Smartphones
Qualcomm Press Release USA Overview Qualcomm has officially launched its Snapdragon X7 Gen 2 platform, designed for AI PCs and premium smartphones, featuring an integrated NPU that doubles its predecessor's AI processing power. This new ... -
Liquid AI Debuts LFM2.5-2.6B: A Powerful AI Agent Model Running on Raspberry Pi Without Cloud or GPUs
VentureBeat USA Overview AI startup Liquid AI debuted LFM2.5-2.6B, a new open-weight language model designed for agentic workloads that can run on local hardware without cloud inference or GPUs. This model enables edge AI applications on... -
Moonshot AI Releases 2.8 Trillion Parameter Kimi K3 MoE Model, NVIDIA Introduces 4B-Parameter Cosmos 3 Edge for On-Device AI
AI Weekly USA Overview Moonshot AI officially released Kimi K3, a 2.8-trillion-parameter Mixture-of-Experts model with a 1-million-token context window and native vision capabilities. In parallel, NVIDIA introduced Cosmos 3 Edge, a 4-bil... -
Apple Unveils Next-Gen ‘Apple Intelligence’ and Smarter ‘Siri AI’ at WWDC26, Integrated Across iOS, iPadOS, and macOS
Apple (Press Release) USA Overview At WWDC26, Apple announced the introduction of next-generation 'Apple Intelligence' and a significantly smarter, more knowledgeable, and capable 'Siri AI,' deeply integrated across its operating systems... -
Google AI Edge Achieves 19.6ms Low-Latency On-Device AI with LiteRT GPU Accelerator on Samsung Galaxy S24
Google AI Edge USA Overview Google AI Edge announced advancements in its LiteRT GPU Accelerator, which, while not yet open-sourced, is available as prebuilts for Kotlin and C++ SDK users. Benchmark results on a Samsung Galaxy S24 device ...
1