Ollama reduced local model hosting from an afternoon of CUDA troubleshooting to a single pull command. It handles model download, quantisation selection, GPU offloading, and memory management, and exposes an OpenAI-compatible endpoint on localhost.
Inference engine that runs LLMs on almost any hardware
Stars126.6K
Commit8h ago
LicenseMIT
llama.cpp is the layer most local model tooling is built on. It is a C and C++ inference engine with no Python runtime, aggressive quantisation support, and hand-tuned kernels for a remarkably wide range of hardware.
Speech recognition that runs anywhere, with no runtime
Stars43.2K
Commit1d ago
LicenseMIT
whisper.cpp is a C and C++ port of the Whisper speech recognition model with no Python dependency, quantised weights, and hardware acceleration across Apple Silicon, CUDA, and plain CPU.
Open-source local AI meeting notes for macOS and Windows
Stars31.6K
Commitjust now
LicenseMIT
Meetily is an MIT-licensed, open-source meeting assistant for macOS and Windows that transcribes on-device and drafts minutes, summaries, and action items privately.
Free open-source macOS dictation with on-device AI cleanup
Stars11.9K
Commitjust now
LicenseGPL-3.0
FluidVoice is a free macOS dictation app that transcribes speech on the machine and types polished text into any app, positioned as a local alternative to Wispr Flow.
VoiceInk transcribes speech on the machine itself using a local Whisper model, then runs the raw transcript through a cleanup pass that removes filler and applies punctuation.
Private offline AI chat with on-device LLMs for iPhone and Mac
Stars2,267
Commit1y ago
LicenseMIT
Fullmoon is a free app that chats with small open language models running entirely on your iPhone, iPad, or Mac, with no account and nothing sent to the cloud.
Offline voice-to-text dictation with a one-time lifetime license
Stars732
Commitjust now
LicenseAGPL-3.0
Voicetypr is an offline-first voice-to-text dictation app for macOS and Windows that transcribes speech on the device and types the result wherever the cursor is, sold as a one-time $49 lifetime license.
Desktop application for running language models locally
CompanyElement Labs
Since2023
PricingFree
LM Studio made local models approachable for people who do not want a terminal. Browse a catalogue, download a quantisation that fits the machine, and chat, with hardware requirements explained in plain language.