@divnatech.bsky.socialLogitech представи Rally Bar 2 с 4 пъти повече Edge AI мощност, Rally System 2 Pro за големи зали и Sound Shape за равномерно аудио покритие до 20 метра.
Rally Bar 2 струва 4299 евро.
#Logitech #EdgeAI #EnterpriseIT #HybridWork
@futuregearai.bsky.socialA new edge AI architecture (AO-DA) splits persona and logic into separate inference paths on a single INT4 base model with hot-swappable LoRA adapters, aiming to keep on-device LLM reasoning robust against persona-heavy…
#EdgeAI #OnDeviceLLM #AIHardware #LLMAgents
https://arxiv.org/abs/2610.09772
@futuregearai.bsky.socialRRAM analogue content-addressable memory enables metric-based classification and online adaptation for extreme edge AI, using configurable receptive field neurons with radial basis activation on a custom Template piXeL cell. Hardware…
#EdgeAI #RRAM #NeuralHardware
https://arxiv.org/abs/2606.14739
@robotcurrent.bsky.socialA study across two real Home Assistant setups finds that small 0.8B–4B models can match frontier models on easier agentic tasks, while capability rankings shift by call site. Larger models are not uniformly better, and one 4B…
#Robotics #HomeAutomation #LLM #EdgeAI
https://arxiv.org/abs/2610.09021
@aidailypost.comMicrosoft just opened pre‑orders for the $5,999 Surface AI Dev Box—powered by Nvidia RTX Spark, ready for edge AI, local compute & next‑gen AI agents. Want to see how it could speed up your models? Dive in! #SurfaceRTXSpark #EdgeAI #AIagents
@edgeai.bsky.social🔥 APPLY FOR **FREE HARDWARE** BUILD SOMETHING THAT MATTERS.🔥
10 days left to submit an idea on how #EdgeAI helps communities prepare for wildfires, floods, and infrastructure failures 😍 50 semi-finalists receive an Arduino UNO™ Q #devkit
https://deveco.io/contests/qualcommforamoreresilientamerica
@aiadoptionagency.bsky.socialMiniCPM V 4.6 runs visual AI locally on phones and servers. No cloud. No data leaks. Reads images, video, and documents. Field service, retail AI, document processing, video knowledge capture. Full guide: aiadoptionagency.com/minicpm-v-4-6/ #EdgeAI #VisionAI #LocalAI
@gothartech.bsky.socialThe fastest inference call is the one that never crosses an ocean. Edge AI moves quantized models to the CDN layer -- Cloudflare Workers AI, Deno, Vercel -- placing intelligence at the same tier as your static assets.
Edge AI Inference: Running Models at the CDN Layer
@futuregearai.bsky.socialA new framework compresses SSL-based audio deepfake detectors from 300M+ parameters down to 31.9M while achieving 6.3× FLOPs reduction, enabling deployment on smaller hardware.
#AIresearch #EdgeAI #AudioSecurity #ModelOptimization
https://arxiv.org/abs/2610.05264
@ai-bloom.warp-studio.comEmbeddingGemma 2：テキスト・画像・音声・動画を統合するオープンな軽量マルチモーダル埋め込みモデル
Google DeepMindが発表した「EmbeddingGemma 2」は、テキスト・画像・音声・動画を単一の空間にマッピングする7.4億パラメータのオープンな軽量マルチモーダル埋め込みモデルです。
@promptfoundry.bsky.socialA new pre-deployment framework estimates task difficulty for federated perception systems by combining intrinsic data properties with client-distribution composition. Experiments on three MNIST variants show strong negative…
#EdgeAI #FederatedLearning #AIResearch
https://arxiv.org/abs/2603.28282
@futuregearai.bsky.socialRepTC introduces a representation-aware strategy that co-optimizes model architecture, session length, and header preprocessing for traffic classification on constrained IoT edge devices. The approach targets joint control of…
#IoT #EdgeAI #TrafficClassification
https://arxiv.org/abs/2610.04784
@futuregearai.bsky.socialSplit Computing for DNN inference faces combined hardware fault and adversarial risks, with a new paper proposing the Joint Vulnerability Score to assess both together. Relevant for edge AI device reliability.
#SplitComputing #EdgeAI #AdversarialML #DNNRobustness
https://arxiv.org/abs/2610.04396
@lunar.computerTechNexion TELOS-AI4000 Packs 40 TOPS of Edge AI With Open-Source Linux Tooling
@thedailytechfeed.com19-yr-old raises $11M to build $3,499 AI agent PC with local data storage. #AI #Hardware #EdgeAI #Privacy #PersonalAI #Startups thedailytechfeed.com/19-year-old-...
@thedailytechfeed.comSpatial models, privacy AI, edge chips: five PearX startups showing what’s next in the mid-stage acceleration world. #AI #EdgeAI #Privacy #3DModeling #TrustTech #PearX thedailytechfeed.com/5-pearx-star...
@hacks.grAdapting AI to a small chip can take time: Lola Vision’s founder says manual preparation may require about 200 hours before testi…
@thedailytechfeed.comLola Vision speeds up AI-on-chip deployment with its new compiler toolchain and chip plans. #EdgeAI #AICompilers #Semiconductors #Startup #MissionCriticalAI #LolaVision thedailytechfeed.com/lola-vision-...
@securityonline.bsky.socialThe Apple smart home camera reportedly records no video, using on-device AI and face recognition to send text alerts. Launch may be October 13.
#Apple #SmartHome #HomeKit #Siri #Privacy #EdgeAI #HomePodMini #AppleTV