Run models locally
Download open models and run them on your own Mac.
Open sourceOllamaOllamaRun open models with one command or in its Mac app; its local API on port 11434 is what many other apps expect.
ViewRuns on your Macggml-orgllama.cppThe C/C++ engine behind much of local AI: runs GGUF models with Metal on Apple silicon, with a web chat and local API server.
View
ViewRuns on your Macggml-orgllama.cppThe C/C++ engine behind much of local AI: runs GGUF models with Metal on Apple silicon, with a web chat and local API server.ViewAll Run locally
Best MLX Score first.
12 of 12
1OllamaRun open models with one command or in its Mac app; its local API on port 11434 is what many other apps expect. 4.8View2llama.cppThe C/C++ engine behind much of local AI: runs GGUF models with Metal on Apple silicon, with a web chat and local API server. 4.6View3
LM StudioA Mac app to find, download and chat with open models, running MLX and GGUF builds, with a local OpenAI-style server. 4.3View4
Open WebUIA self-hosted chat interface in your browser for Ollama and OpenAI-compatible models, local or in the cloud. 3.9View5
AnythingLLMA desktop AI assistant to chat with your documents, run agents and transcribe meetings, with local or cloud models. 3.4View6mlx-lmApple's MLX tools for language models: generate, chat, convert, quantize and fine-tune models from the command line. 2.9View7
exoSplits a model across several Macs on your network so they run it together as one local AI cluster. 2.6View8LocalAIA self-hosted server with an OpenAI-compatible API for text, speech, images and more, using engines like llama.cpp and MLX. 2.5View9JanAn open-source chat app that runs models on your Mac with a bundled llama.cpp engine, with a local API and optional cloud models. 2.2View10
LlamaA small menu bar app from the llama.cpp team that installs models suited to your Mac and serves them on a local API. 2.1View11
vMLX (MLX Studio)A Mac app and MLX inference engine for chat, coding tools, image generation and an OpenAI- or Anthropic-style local API. 1.1View12Msty StudioA desktop AI workspace for chatting with local and online models, with knowledge stacks, personas and prompt libraries. 1.0View
OllamaRun open models with one command or in its Mac app; its local API on port 11434 is what many other apps expect. 4.8View2llama.cppThe C/C++ engine behind much of local AI: runs GGUF models with Metal on Apple silicon, with a web chat and local API server. 4.6View3LM StudioA Mac app to find, download and chat with open models, running MLX and GGUF builds, with a local OpenAI-style server. 4.3View4Open WebUIA self-hosted chat interface in your browser for Ollama and OpenAI-compatible models, local or in the cloud. 3.9View5AnythingLLMA desktop AI assistant to chat with your documents, run agents and transcribe meetings, with local or cloud models. 3.4View6mlx-lmApple's MLX tools for language models: generate, chat, convert, quantize and fine-tune models from the command line. 2.9View7exoSplits a model across several Macs on your network so they run it together as one local AI cluster. 2.6View8LocalAIA self-hosted server with an OpenAI-compatible API for text, speech, images and more, using engines like llama.cpp and MLX. 2.5View9JanAn open-source chat app that runs models on your Mac with a bundled llama.cpp engine, with a local API and optional cloud models. 2.2View10LlamaA small menu bar app from the llama.cpp team that installs models suited to your Mac and serves them on a local API. 2.1View11vMLX (MLX Studio)A Mac app and MLX inference engine for chat, coding tools, image generation and an OpenAI- or Anthropic-style local API. 1.1View12Msty StudioA desktop AI workspace for chatting with local and online models, with knowledge stacks, personas and prompt libraries. 1.0View