Local AI

rt-detr implementation in C++ with ggml

A from-scratch C++17/ggml port of Depth Anything 3 (ByteDance)

Adaptive Precision for EXpert Models: MoE-aware mixed-precision quantization

Animate portrait photos with your voice - presently implements FLOAT

Shared Go client for pairing applications and services with the Dante platform

Lean neural real-time acoustic echo cancellation with soft delay estimation - GGML and PyTorch inference

moss-transcribe.cpp is a from-scratch C++17 inference port of OpenMOSS MOSS-Transcribe-Diarize, built on ggml

C++ port of Microsoft VibeVoice built on ggml

OpenAI's privacy filter NER model architecture implemented in a minimal C++/GGML runtime