edge-ai repositories

Trending repositories tagged edge-ai, ranked by trending score. Updated daily.

  1. 1
    Edge0-AI/Edge0

    Device-native foundation models: 145B on a laptop, 35B on a phone. Latent MoE, SSD-streamed weights, a router that predicts loads a step ahead. Mac, iPhone, Android, Windows.

    GitHub · C++ · 12,180 stars · 1,885 forks · +127 stars in 24 hours · Trending score 4.61

  2. 2
    hyperionaidev/smolmaxxing

    maxxing the ability to run different LLMs on the smallest devices possible

    GitHub · C · 1,083 stars · 61 forks · +64 stars in 24 hours · Trending score 3.69

  3. 3
    Intellindust-AI-Lab/GTR

    GTR: Gated Token Recurrence for Efficient Dense Prediction

    GitHub · Python · 150 stars · 14 forks · +99 stars in 24 hours · Trending score 3.44

  4. 4
    0xShug0/audio.cpp

    An all-in-one, pure C++ inference engine for audio models, powered by ggml. Supports TTS, STT, VAD, voice conversion, music generation, and more, with highly optimized performance. No Python dependency.

    GitHub · C++ · 4,386 stars · 545 forks · +33 stars in 24 hours · Trending score 3.20

  5. 5
    cactus-compute/cactus

    Sub-50mb agentic, voice and vision intelligence for consumer devices.

    GitHub · Python · 14,866 stars · 1,046 forks · +41 stars in 24 hours · Trending score 3.18

  6. 6
    Helldez/BigMoeOnEdge

    Run MoE models bigger than your RAM. Frontier-size MoE on a phone, laptop, edge. Lossless, on stock llama.cpp. An adaptive streaming optimizer. Any hardware, any OS, any model, any quantization, behind a local OpenAI-compatible server.

    GitHub · C++ · 1,066 stars · 111 forks · +26 stars in 24 hours · Trending score 3.16

  7. 7
    Ampixa/sanoTTS

    sanoTTS (सानो = 'small' in Nepali): 294k–2.27M-parameter neural TTS that runs on a $3 chip, in the browser, or in Python with low-latency streaming.

    GitHub · C · 851 stars · 90 forks · +60 stars in 24 hours · Trending score 3.10

  8. 8
    AbrahamPaulJ/facefusion-mobile

    FaceFusion for Android. Offline face swap running on the phone's NPU, GPU or CPU. Qualcomm Hexagon (v68 to v81), Vulkan fallback, no server and no upload.

    GitHub · Kotlin · 145 stars · 77 forks · +41 stars in 24 hours · Trending score 2.93

  9. 9
    nats-io/nats-server

    High-Performance server for NATS.io, the cloud and edge native messaging system.

    GitHub · Go · 21,092 stars · 2,052 forks · +21 stars in 24 hours · Trending score 2.62

  10. 10
    lokutor-ai/ito

    Natural-sounding streaming text-to-speech for the ESP32-S3: 3.3 M params, 3.8 MB of weights, no cloud, no NPU

    GitHub · C · 38 stars · 2 forks · +26 stars in 24 hours · Trending score 2.50

  11. 11
    AbrahamPaulJ/nightmare-mobile

    On-device, ComfyUI-style node graph for image/video diffusion on Android. Run MiniMax H3 and WAN 2.2 14B on the Qualcomm Hexagon NPU. SD1.5, SDXL, Flux, Z-Image, Anima, Qwen Image 2.1, Krea 2 & MOSS Image. Inpainting, outpainting, ControlNet, LoRA, IP-Adapter & LivePortrait. Fully offline, no server, no account.

    GitHub · Kotlin · 290 stars · 24 forks · +19 stars in 24 hours · Trending score 2.48

  12. 12
    ultralytics/yolov3

    PyTorch implementation of YOLOv3, YOLOv3-SPP, and YOLOv3-tiny for real-time object detection with training, validation, inference, and multi-format export.

    GitHub · Python · 10,647 stars · 3,407 forks · +31 stars in 24 hours · Trending score 2.26

  13. 13
    Renakoni/Voicebox

    最好用的 Android 离线语音克隆与三语 TTS(误)——基于 GPT-SoVITS,支持中日英。 | The best offline voice cloning & trilingual TTS on Android (jk) — powered by GPT-SoVITS (ZH/JA/EN).

    GitHub · Kotlin · 208 stars · 42 forks · +87 stars in 24 hours · Trending score 2.17

  14. 14
    google-ai-edge/LiteRT-LM

    LiteRT-LM is Google's production-ready, high-performance, open-source inference framework for deploying Large Language Models on edge devices.

    GitHub · C++ · 6,913 stars · 798 forks · +9 stars in 24 hours · Trending score 2.12

  15. 15
    Desert-Ant-Labs/desert-ant-core

    On-device AI SDKs for iOS, macOS, Android, and the web. Small, focused models that run fully offline in Swift, Kotlin, and JavaScript with Core ML, LiteRT, and WebAssembly.

    GitHub · Swift · 463 stars · 27 forks · +15 stars in 24 hours · Trending score 2.10

  16. 16
    off-grid-ai/OGAM

    The Swiss Army Knife of Offline AI. Chat, see, speak, and generate images on your phone or Mac — GGUF LLMs, vision, Whisper speech-to-text, Stable Diffusion, tool calling, and local-network servers. Runs on your CPU, GPU, or NPU. No account, no API key, zero data leaves your device.

    GitHub · TypeScript · 3,407 stars · 341 forks · +2 stars in 24 hours · Trending score 1.82

  17. 17
    Z-Teddy/Gimbal-AI-Assistant

    基于 RK3588 + STM32 的异构嵌入式智能云台系统,支持视觉感知、模式闭环、主动搜索、UART 通信、本地语音交互与本地 LLM 对话。

    GitHub · C · 33 stars · 11 forks · +8 stars in 24 hours · Trending score 1.68

  18. 18
    ThinkOffApp/CarWatch

    Your car as a chat-room agent: Raspberry Pi 5 + dashcam + local AI. CodeWatch's sibling for the garage.

    GitHub · Python · 306 stars · 20 forks · +10 stars in 24 hours · Trending score 1.66

  19. 19
    beclab/Olares

    Open-Source Personal Cloud OS for Always-On Agents

    GitHub · Go · 5,373 stars · 340 forks · +5 stars in 24 hours · Trending score 1.63

  20. 20
    afjcjsbx/picoclaw

    Ultra-lightweight, open-source, self-hosted personal AI assistant in Go for RISC-V, SBCs, and embedded systems. Active PicoClaw fork with WebUI, MCP, memory, multi-agent workflows, plugins, and chat apps.

    GitHub · Go · 62 stars · 18 forks · +5 stars in 24 hours · Trending score 1.62

  21. 21
    microsoft/SPARROW

    SPARROW — Solar-Powered Acoustic and Remote Recording Observation Watch. An AI-enabled edge device for biodiversity monitoring with camera-trap imaging and bioacoustic recording. By Microsoft AI for Good Lab.

    GitHub · Python · 73 stars · 22 forks · +5 stars in 24 hours · Trending score 1.47

  22. 22
    lotvulture/lotvulture

    Real-time parking lot occupancy detection using existing security cameras and on-device computer vision.

    GitHub · Python · 189 stars · 32 forks · −2 stars in 24 hours · Trending score 1.34

  23. 23
    thebasedcapital/ane-infer

    Apple Neural Engine (ANE) LLM inference engine — reverse-engineered private APIs, Metal GPU shaders, hybrid ANE+GPU+CPU on Apple Silicon. 32 tok/s matching llama.cpp, 3.6 TFLOPS fused ANE mega-kernels.

    GitHub · Rust · 35 stars · 5 forks · +6 stars in 24 hours · Trending score 1.28

  24. 24
    DEEPX-AI/dx_rt_npu_linux_driver

    An open source npu linux PCIe driver for interacting with DEEPX devices over the PCIe interface

    GitHub · C · 15 stars · 9 forks · +4 stars in 24 hours · Trending score 1.17