LLM
A command-line tool and Python library for running prompts against OpenAI, Anthropic Claude, Gemini and dozens of other language models, including local models, with SQLite logging and plugin support.
LLM is a command-line tool and Python library for interacting with OpenAI, Anthropic's Claude, Google's Gemini, and dozens of other large language models, both through remote APIs and with models installed and run locally. It can run prompts from the command line, store prompts and responses in a local SQLite database, generate and store embeddings, extract structured content from text and images, and give a model the ability to call tools.
It installs with pip, Homebrew, pipx or uv, and can run prompts or chats against any OpenAI-compatible endpoint, such as a locally running LM Studio server, without any extra configuration. Support for individual providers and local runtimes comes through a plugin system: separate plugins add access to providers such as Gemini and Anthropic directly, or to models running locally through Ollama, each managed with its own API key or local setup.