The Offline AI Modules workstream's first end-to-end benchmark evaluation found that Q4_K_M quantization offers the best size-to-quality trade-off for offline voice-first AI deployment, with gemma-4-E…
A developer built FloraTrail AI, a fully offline, screenless voice assistant for hikers that runs on a Raspberry Pi 5 or pocketable Linux/Android device. The system combines whisper.cpp for on-device …
A developer built Kaya, a construction-site safety copilot combining a Raspberry Pi 5 headset with YOLO11, YOLO-World, YOLO-Pose, Depth Anything V2, a PPE checker, a Docling-based RAG index, and Gemin…
Cactus Compute released Needle 2, a 14MB function-calling LLM that runs entirely on a Raspberry Pi 5's CPU and selected the correct tool for the prompt "Turn the LED on" in 78 milliseconds. The model,…
WIM-Z, a tracked pet-care robot built on a Raspberry Pi 5 with a Hailo-8 AI accelerator, reported barking reductions of up to 83% across seven field trials run between May and September 2026, with fou…
A developer released anything2explainer, an open-source Claude Code and Codex agent skill that turns a topic, article or document into a narrated explainer video. Every frame is drawn with Remotion in…
Cactus Compute released Needle 3 on September 17, an Apache 2.0-licensed automation foundation model that ships as an 8 to 29MB binary and runs tool calling, structured extraction, and text embeddings…
A Halloween costume detector project uses a Raspberry Pi 5 with a NoIR PiCamera and speaker bonnet to run a Python script with OpenCV person detection, sending pictures to Claude for costume recogniti…
A new e-paper display project by [arnegiacomo] uses a Raspberry Pi 5 and BirdNet-Go, an AI-powered birdsong classifier, to automatically draw pictures of birds heard outside in real time, assembling i…
VM-1, a Raspberry Pi 5 based videomixer and videoplayer, has been released as a DIY project by developers Nils and Julian, originating from the digital lab at the Ernst Busch University of Theatre Art…
A custom llama.cpp backend (ggml-axcl) now runs Qwen3-0.6B directly from GGUF files on an Axera AX8850 NPU accelerator card (M5Stack LLM-8850) hosted on a Raspberry Pi 5, achieving 1.3-2.7 tokens per …
A developer built CarWatch, a Raspberry Pi 5-based in-car AI assistant that runs a 35B-parameter Qwen3.6-35B-A3B model locally at 3.5 tokens per second, achieving full offline functionality for voice,…
A Hackernoon guide details how to self-host AI models on a Raspberry Pi 5, citing costs of $40-60 per month on OpenAI API calls for an agent project. The guide explains the setup process and benefits …
Crewai-go v0.4.0, a Go implementation of the CrewAI multi-agent orchestration framework, has been released, offering type safety, lower memory footprint, and faster concurrent execution compared to th…
Google engineers built Gemma Translator, an open-source project that runs a fully offline multilingual voice translator on a Raspberry Pi 5 using Google Gemma 4 E2B (2.3B effective / ~5.1B total param…
Cactus Compute released Needle 2, an open 45M-parameter tool-calling model that ships as a 14MB binary and runs a full session in about 28MB of RAM, targeting devices with no GPU or NPU. The model ach…
Google AI Edge's LiteRT and LiteRT-LM enable deployment of Gemma 4 E2B on Raspberry Pi 5, achieving 99 tokens/sec prefill and 9 tokens/sec decode with a peak memory footprint of 1432 MB, powering the …
Google's Gemma Translator, built with Google Antigravity, enables fully offline, on-device voice translation using the Gemma 4 large language model and LiteRT-LM, with text-to-speech powered by Moonsh…
Cactus Compute released Needle2, a 45-million-parameter language model compressed to a 14MB binary that runs AI agent tool-calling at 500 tokens per second on a Raspberry Pi 5 and 6,000 tokens per sec…
Cactus released Needle 2, a 14MB agentic LLM for tool call, device use, and structured extraction, running 45 million parameters at 2-bit compression in 28MB of RAM. It achieves 500 tokens/sec on a Ra…