NNakonakokawaiiinai-change-brief.hashnode.dev·13h ago · 9 min readWhat I Learned While Exploring Local AI ToolsRecently, I have been spending more time exploring local AI tools and learning how different parts of the ecosystem work together. Until recently, I mostly thought about AI through web services and ch00
NNakonakokawaiiinai-change-brief.hashnode.dev·21h ago · 5 min readGemma 4 12B on a 16GB Laptop: A Privacy-First Local AI Business PlaybookGoogle’s mid-sized multimodal open model creates a credible path to on-device services, but the winning offer is controlled workflow design—not a vague promise that “local” solves everything. For a sm00
BKBrian Kinginsolodev.app·4d ago · 12 min readInstalling vLLM onto EVO-X2.Abstract. I download a vLLM Docker image that supports AMD, and the ROCm GPU driver that runs on the Strix Halo platform (AI Max+ 395 APU). I connect the running Docker container to local GGUF AI mode00
RCRobert Collinsinselfenrichment.hashnode.dev·Jul 1 · 8 min readdgx-spark-inference - Keeping good habits around local inferenceLocal inference has a habit of becoming folklore. You launch a model from a shell history fragment you can never quite find again. A context setting lives in a note somewhere. A faster quantization ge00
TStarini sunilinai-content-utilties.hashnode.dev·Jun 22 · 4 min readWhat Your Documents Mean, Not Just What They SayOn this page A Tiny Experiment Why Keywords Fail Meaning Without Matching Words The Hidden Layer A Simple Mental Model Try It Yourself The Aha Moment The Hidden Signal In the previous artic00
NCNube Colectivainnubecolectiva.hashnode.dev·Jun 22 · 5 min read6 Free Local AI Models That Are Getting Close to Claude in 2026For years, models like Claude became the reference point for developers, writers, researchers, and professionals who needed strong reasoning, long-context analysis, coding ability, and natural convers00
JWJiyao Wenginwengjiyao.hashnode.dev·Jun 14 · 12 min readGemma 4 for Telephony: I Replaced Two AI Models With One in My Voice Phone Agent — Until I Switched to ChineseBuilding a phone agent on a multimodal LLM: dropping faster-whisper and letting Gemma 4 hear the caller directly — a response-time and reply-accuracy benchmark across English, French, and Mandarin My 00
YPYaroslav Pristupaincode-friendly.hashnode.dev·Jun 8 · 12 min readStep-by-step guide to running Gemma-4 26B locally on budget GPUsThe local AI revolution just got a serious upgrade. Google's Gemma-4 26B model, combined with Unsloth's Quantization-Aware Training GGUF formats, makes it possible to run a 26-billion parameter model 00
JCJim Chundevalelinaccidentalcomplexity.hashnode.dev·Jun 1 · 18 min readPart 1: Building a Self-Hosted AI Agent Cluster with Raspberry Pi 5 and Hailo AI HAT+ 2Introduction So there I was, staring at four Raspberry Pi 5s, two Hailo AI HAT+ accelerators, a growing pile of cables, and the uncomfortable realization that I had built a tiny data center to avoid s00
TStarini sunilinai-content-utilties.hashnode.dev·May 30 · 5 min readWhat Your Documents Whisper When Nobody's LookingMost people think important information is easy to spot. Look at enough documents and you will find the important topics because they appear again and again. More mentions = more importance. Right? No00