# yolo_ggml

**Repository Path**: cumt/yolo_ggml

## Basic Information

- **Project Name**: yolo_ggml
- **Description**: command line app for main development of GGML/YOLO app
- **Primary Language**: C++
- **License**: MIT
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2026-09-24
- **Last Updated**: 2026-09-24

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# yolo_ggml
command line app for main development of GGML/YOLO app

# build

- `git clone https://github.com/ayaromenok/yolo_ggml.git`
- `cd yolo_ggml`
- `mkdir build`
- `cd build`
- `cmake -DGGML_BACKEND=VULKAN ../`

BackEnd can be `CUDA`, `CPU`, `VULKAN`, `OPENCL`

# Hardware

## TinkerBoard

With latest Armbian/trixier it's no Vulkan/OpenCL acceleration, so CPU only and 32bit

 | model | size, MB | perf, ms |
 |-------|----------|----------|
 | 26n   |  4.99    |  240     |
 | 26s   | 19.21    |  491     |
 | 26m   | 41.96    | 1228     |

# YOLO GGML Android
YGA - is native android application to work with YOLO models with a help GGML library. Use GPU acceleration via Vulkan backend for now