* [Off Grid AI v0.0.111](https://github.com/off-grid-ai/OGAM) – Offline on-device AI suite for chat, image generation, vision, speech-to-text, and tool calling across mobile and macOS, using local GGUF and hardware acceleration.

* [Semantic Search v0.5.0](https://github.com/kelindar/search) – Go library for embedded vector search and GGUF BERT-based semantic embeddings with optional GPU acceleration.

* [AJEAN v0.18.0](https://github.com/nathaninline/ajean) – AJEAN is a home AI assistant single-binary setup with chat, persistent memory, tools, web and browser control, and end-to-end encrypted remote access.

* [atomic-agent v0.6.7](https://github.com/AtomicBot-ai/atomic-agent) – Local operator agent for llama.cpp that controls browser, files, shell, tasks, and approvals while keeping data and models local.

* [ReadyLLM v0.3.0](https://github.com/htobty/ReadyLLM) – Visual interface for deploying, monitoring, and tuning local LLMs, with hardware-aware parameter search and ComfyUI-based image and video generation.

* [ggrun v3.2.10](https://github.com/raketenkater/ggrun) – Auto-tuned launcher that measures multi-GPU hardware for GGUF models, picks an optimal llama.cpp/ik\_llama.cpp backend, and serves an OpenAI-compatible API.

* [Gerbil v1.28.4](https://github.com/lone-cloud/gerbil) – Desktop app for running Large Language Models locally with cross-platform support and integrated image generation.

* [LLMKube llmkube-0.10.2](https://github.com/defilantech/LLMKube) – Kubernetes operator managing self-hosted LLM inference on NVIDIA GPUs and Apple Silicon, with autoscaling, model routing, and OpenAI-compatible API.

* [tensai v0.0.35](https://github.com/mattn/tensai) – Tiny machine-learning framework for Go, implementing forward passes, backpropagation, and optimization with optional SIMD and WebGPU acceleration.

* [Off Grid AI v0.0.51](https://github.com/off-grid-ai/OGAD) – Local-first AI runtime with a studio and an OpenAI-compatible gateway, running open multimodal models entirely on-device with an optional always-on assistant layer.