Keep Calm and Read the Friendly Manual :-)
Description:
whisper.cpp is a high-performance C/C++ port of OpenAI’s Whisper for offline transcription. It runs on Apple Silicon with NEON/Metal and on x86 with AVX, supports quantization, Vulkan and multiple backends, and includes a lightweight ggml-based engine for cross-platform use (macOS, iOS, Linux, Windows, WASM).
Description:
Burn is a next-generation tensor library and Deep Learning Framework, designed for flexibility, efficiency, and portability. Built in Rust, it offers dynamic graphs with static-graph performance via JIT, multiple backends (CPU, GPU, WebAssembly), ONNX import, and training to deployment across devices.
Description:
OpenClaw is a personal AI assistant that runs on your own devices across any OS and platform, with multi-channel support and extensible skills, designed to keep data on-device.
Description:
PersonaPlex is a real-time, full-duplex speech-to-speech model that enables persona control via text prompts and voice conditioning. Trained on synthetic and real conversations, it delivers natural, low-latency interactions and runs with the Moshi architecture; the code is MIT-licensed, with model weights released under NVIDIA’s Open Model license.
Description:
Simba is an open-source, portable Knowledge Management System designed for retrieval-augmented generation (RAG). It helps manage, process, and retrieve knowledge from documents to enhance AI applications, featuring a Python SDK, modular architecture, a modern UI, and an embeddable Simba chat widget; a self-hosted, customizable solution.
Description:
Apollo-Lite provides powerful modules and features for autonomous driving development.
Description:
KoboldCpp is a single-file, install-free AI text-generation app for GGML and GGUF models, inspired by KoboldAI. It runs on CPU or GPU, includes a KoboldAI Lite UI, supports image generation, Whisper recognition, and TTS, with ready-made binaries for Windows, macOS, and Linux.
Description:
Gerbil is a cross-platform desktop app for running Large Language Models locally, handling backends, model downloads, and hardware acceleration, with built-in image generation, HuggingFace model search, and optional OpenWebUI/SillyTavern integrations.
Description:
AMD’s open-source framework for building AI agents that run locally on AMD Ryzen AI hardware with privacy, no cloud costs, and hardware acceleration.
Description:
Connect your devices into an AI cluster to run large AI models across multiple devices with automatic discovery and topology-aware parallel processing