Edge Computing News Edge computing coverage at The Robotics Media tracks how on-device processing powers robots, drones, and autonomous systems—reducing latency, bandwidth use, and cloud dependence. Readers find news on edge AI chips and accelerators, embedded platforms from vendors like NVIDIA, Qualcomm, and Intel, real-time inference frameworks, and 5G-enabled deployments. Coverage spans industrial automation, autonomous vehicles, and smart infrastructure, including product launches, benchmarks, and research developments.
BrainChip Puts Akida AKD1500 On A PCIe Card So Any Desktop Can Run Edge AI — Sep 26, 2026 — BrainChip drops its Akida AKD1500 neuromorphic silicon on a plug-and-play PCIe card, aiming to move edge-AI evaluation from lab benches to any worksta...PrismML Ports 1-Bit Bonsai LLM To Qualcomm Snapdragon Smart Glasses — Sep 26, 2026 — PrismML, an AI lab founded by Caltech researchers and advised by UC Berkeley's Ion Stoica, unveiled a Qualcomm Snapdragon AR1 Gen 1 build of its 2B-pa...QBit Semiconductor Targets Humanoids And Drones With QB88XX 'Cerebellum' Chip — Sep 18, 2026 — Taipei-listed QBit Semiconductor unveiled a Cerebellar Model Articulation Controller SoC that runs a 32-motor humanoid hand on one chip, plus a QB77XX...Huawei Unveils Atlas 960 SuperPoD, Pulls Ascend 960 In Three Quarters Early — Sep 17, 2026 — Huawei unveiled the Atlas 960 SuperPoD at Huawei Connect 2026 with 4,096 Ascend 960 accelerators, 8 exaflops FP8 and NPO interconnects, pulling the As...TOYO Buys A Second IQM Machine, Adding IQM Spark To Japan's Quantum Test Bench — Sep 17, 2026 — Tokyo-listed technical house TOYO Corporation is buying its second IQM quantum computer — the 5-qubit IQM Spark — for its Kiba R&D centre, alongside t...NVIDIA opens CUDA-Q Logical for fault-tolerant quantum — Sep 16, 2026 — NVIDIA has extended its CUDA-Q platform with an open-source layer for fault-tolerant quantum computing, adopted at launch by Fermilab, IQM, Infleqtion...Anyon unveils open-source NVQLink quantum control plane — Sep 16, 2026 — Anyon Computing has released an open-source, real-time control system that treats QPUs, CPUs and GPUs as one supercomputer over NVIDIA's NVQLink fabri...Meta Puts MTIA v3 'Iris' Custom AI Chip Into Production This Month — Sep 11, 2026 — Meta has begun production of its third-generation MTIA 'Iris' AI accelerator this September, a Broadcom-designed, TSMC 3nm chip meant to double the co...IBASE Debuts PI900 Pico-ITX Board With Intel Core Ultra For Edge AI — Sep 5, 2026 — Taiwan's IBASE Technology has launched the PI900 Pico-ITX motherboard, packing Intel Core Ultra CPU, GPU and NPU acceleration plus 48GB DDR5 into a 10...D-Robotics Puts Sunrise Chips Under IFA 2026's Buzziest Home Robots — Sep 4, 2026 — D-Robotics used the IFA 2026 opening to showcase its Sunrise embodied-AI chips inside a new wave of home robots, including TCL's hey AiMe companion ro...Perplexity Ships Hybrid AI That Keeps Sensitive Work On Apple Silicon Macs — Sep 2, 2026 — Perplexity's Computer agent now hands sensitive steps to a locally running model on the user's Mac while keeping heavy reasoning in the cloud.Apple Reveals M6, Its First 2nm Chip With Dual 16-Core Neural Engine — Aug 31, 2026 — Apple's M6 debuts on TSMC's 2nm N2 process with a 12-core CPU, 12-core GPU with per-core neural accelerators, and a Dual 16-core Neural Engine for on-...Celona Debuts Orion Agentic Wireless Fabric For Physical AI — Aug 16, 2026 — Celona has launched Orion, an agentic converged wireless platform that fuses private 5G, Wi-Fi 7 and Starlink under one orchestrator, aimed at autonom...OpenAI Runs GPT-5.6 Sol At 750 Tokens/Second On Cerebras In New Ultrafast Tier — Aug 14, 2026 — Cerebras is powering OpenAI's new Ultrafast API tier, running GPT-5.6 Sol at up to 750 output tokens per second and 14x faster than standard processin...Anthropic Builds In-House AI Chip Team To Slash Claude Costs — Aug 14, 2026 — Anthropic has confirmed it is building an in-house AI chip design team led in part by ex-Tesla Dojo hire Clive Chan, targeting custom silicon co-desig...Acrab Debuts 5nm GΞLIX 1 SoC To Run 100B-Parameter Models At The Edge — Jul 27, 2026 — Singapore's Acrab unveiled the 5nm GΞLIX 1 SoC and Agent Box reference system, promising to run 100-billion-parameter models locally with 7.5x prefill...Peridio Ships Avocado OS 1.0, Pitching An 'Arduino Moment' For Physical AI Fleets — Jul 18, 2026 — Peridio has released Avocado OS 1.0, a production-grade embedded Linux runtime that lets a single engineer take a device from a config file to a signe...NVIDIA Rolls Out Jetson T3000 And T2000 To Take Thor Mainstream — Jul 17, 2026 — NVIDIA has introduced Jetson T3000 and T2000 modules, halving the size and power of Thor while packing 865 and 400 FP4 teraflops respectively, and pai...Rambus Unveils DDR5 9600 Chipset To Feed AI Data Center Memory Demand — Jul 16, 2026 — Rambus rolled out a complete DDR5 9600 server RDIMM chipset built around its new 6th-generation RCD06, delivering 20% more memory bandwidth for the ag...AMD Ryzen AI Embedded P100 Enters Production For Edge AI — Jul 15, 2026 — AMD begins volume shipments of its Ryzen AI Embedded P100 Series in July 2026, targeting industrial PCs, autonomous robots and medical imaging devices...