llm

<!-- [[[cog
# README.md is generated from docs/index.md using sphinx_markdown_builder
import tempfile
import subprocess
from pathlib import Path

readme_markdown = ''

with tempfile.TemporaryDirectory() as tmpdir:
    tmp_path = Path(tmpdir)
    # Run: sphinx-build -M markdown ./docs ./tmpdir
    subprocess.run([
        "sphinx-build",
        "-M", "markdown",
        "./docs",
        str(tmp_path)
    ], check=True)
    index_file = tmp_path / "markdown" / "index.md"
    readme_markdown = index_file.read_text(encoding="utf-8")

cog.out(readme_markdown)
]]] -->
# LLM

[![GitHub repo](https://img.shields.io/badge/github-repo-green)](https://github.com/simonw/llm)
[![PyPI](https://img.shields.io/pypi/v/llm.svg)](https://pypi.org/project/llm/)
[![Changelog](https://img.shields.io/github/v/release/simonw/llm?include_prereleases&label=changelog)](https://llm.datasette.io/en/stable/changelog.html)
[![Tests](https://github.com/simonw/llm/workflows/Test/badge.svg)](https://github.com/simonw/llm/actions?query=workflow%3ATest)
[![License](https://img.shields.io/badge/license-Apache%202.0-blue.svg)](https://github.com/simonw/llm/blob/main/LICENSE)
[![Discord](https://img.shields.io/discord/823971286308356157?label=discord)](https://datasette.io/discord-llm)
[![Homebrew](https://img.shields.io/homebrew/installs/dy/llm?color=yellow&label=homebrew&logo=homebrew)](https://formulae.brew.sh/formula/llm)

A CLI tool and Python library for interacting with **OpenAI**, **Anthropic’s Claude**, **Google’s Gemini**, **Qwen**, **Gemma**, **Kimi**, **DeepSeek**, **Mistral**, and dozens of other Large Language Models, both via remote APIs and with models that can be installed and run on your own machine.

Watch **[Language models on the command-line](https://www.youtube.com/watch?v=QUXQNi6jQ30)** on YouTube for a demo or [read the accompanying detailed notes](https://simonwillison.net/2024/Jun/17/cli-language-models/).

With LLM you can:

- [Run prompts from the command-line](https://llm.datasette.io/en/stable/usage.html#usage-executing-prompts)
- [Store prompts and responses in SQLite](https://llm.datasette.io/en/stable/logging.html#logging)
- [Generate and store embeddings](https://llm.datasette.io/en/stable/embeddings/index.html#embeddings)
- [Extract structured content from text and images](https://llm.datasette.io/en/stable/schemas.html#schemas)
- [Grant models the ability to execute tools](https://llm.datasette.io/en/stable/tools.html#tools)
- … and much, much more

## Quick start

First, install LLM using `pip` or Homebrew or `pipx` or `uv`:

```bash
pip install llm
```

Or with Homebrew (see [warning note](https://llm.datasette.io/en/stable/setup.html#homebrew-warning)):

```bash
brew install llm
```

Or with [pipx](https://pipx.pypa.io/):

```bash
pipx install llm
```

Or with [uv](https://docs.astral.sh/uv/guides/tools/)

```bash
uv tool install llm
```

Use LLM to run prompts or start chats against an arbitrary OpenAI-compatible Chat Completions endpoint, su