Local AI ORG

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

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

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

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

C++ port of Microsoft VibeVoice built on ggml

ggml port of CED SOTA-tier audio-tagging model that classifies everyday sounds

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

Turn images into 3D scenes - GGML port of FreeSplatter: Pose-free Gaussian Splatting for Sparse-view 3D Reconstruction

rt-detr implementation in C++ with ggml

voice-detect.cpp is a from-scratch C++17 inference engine for speaker recognition and voice analysis, built on ggml

No description.

Animate portrait photos with your voice - presently implements FLOAT

Open-source models that turn Italian audio into text, locally, on your own hardware, with optional speaker diarization. Brought to you by the LocalAI team.

Scripts/recipes to make agents record short demos

C++/ggml port of TRELLIS.2 stage-1 geometry (image -> watertight mesh)

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

C++/ggml port of TRELLIS.2 stage-1 geometry (image -> watertight mesh)

No description.