llm

CLI for interacting with large language models

TLDR

Chat with default model

$ llm "[prompt]"

Use specific model

$ llm -m [gpt-4o] "[prompt]"

Interactive chat session

$ llm chat

Continue most recent conversation

$ llm -c "[follow up]"

List models (built-in and from plugins)

$ llm models

Pipe input to a model

$ cat [file.txt] | llm "[summarize this]"

Use a system prompt

$ llm -s "[Reply as a pirate]" "[Hi there]"

Set an API key

$ llm keys set [openai]

Install a plugin (e.g. Claude support)

$ llm install [llm-claude-3]

SYNOPSIS

llm [options] [prompt]llm subcommand [options]

llm is a CLI and Python library for interacting with large language models. Out of the box it talks to OpenAI's API; through plugins it also supports Anthropic Claude, Google Gemini, Mistral, local Ollama, llama.cpp / GGUF, and many other providers and self-hosted models.Every prompt and response is logged to a local SQLite database (under ~/.config/io.datasette.llm/ on Linux or ~/Library/Application Support/io.datasette.llm/ on macOS) so previous conversations can be searched, exported, and replayed. The tool also supports prompt templates, system prompts, embeddings, similarity search, and pluggable backends.

PARAMETERS

PROMPT

Text prompt for the model.

-m MODEL, --model MODEL

Model to use (e.g. gpt-4o, claude-3-5-sonnet, llama2).

-c, --continue

Continue the most recent conversation.

--cid ID

Continue a specific conversation by id.

-s PROMPT, --system PROMPT

Set a system prompt.

-t NAME, --template NAME

Use a named prompt template.

-o KEY VALUE, --option KEY VALUE

Pass a model-specific option (e.g. -o temperature 0).

-a PATH, --attachment PATH

Attach a file (image, PDF, audio) to the prompt for models that support attachments.

--no-stream

Disable token streaming and only print the final answer.

--key KEY

Use a specific API key for this call.

SUBCOMMANDS

chat

Open an interactive chat session.

models [list | default name]

List available models or set the default model.

keys [set key]

Manage API keys.

logs

Show or query the persistent log of prompts and responses.

templates

Manage reusable prompt templates.

install package / uninstall package

Install or remove an llm plugin (e.g. llm-claude-3, llm-ollama).

embed, embed-multi, similar

Generate and search text embeddings.

CAVEATS

Cloud models require API keys stored via llm keys set. Conversation logs are stored in plaintext SQLite, so be mindful of secrets in prompts. Provider support beyond OpenAI requires installing additional plugins (e.g. llm install llm-claude-3).

HISTORY

llm was created by Simon Willison to provide unified CLI access to language model APIs and local models, and to make it easy to log and query past prompts.

llm