tsellm 0.1.0a5

Interactive SQLite shell with LLM support

tsellm: Use LLMs in your SQLite queries

tsellm is the easiest way to access LLMs through your SQLite database.

pip install tsellm

You can then use LLMs in your SQL queries. You can supply your queries through a simple CLI.

tsellm my.sqlite3 "select prompt(p, 'orca-2-7b') from my_table"

You can also enter an interactive REPL shell and try things out.

tsellm my.sqlite3

Examples

Let's create a simple SQLite database with some data.

sqlite3 prompts.db <<EOF
CREATE TABLE [prompts] (
   [prompt] TEXT
);
INSERT INTO prompts VALUES('hello world!');
INSERT INTO prompts VALUES('how are you?');
INSERT INTO prompts VALUES('is this real life?');
INSERT INTO prompts VALUES('1+1=?');
EOF

CLI

You can use any of the llm plugins

You can execute LLM-powered SQL queries directly in the CLI, like so:

First, let's install a dummy plugin.

llm install llm-markov

tsellm prompts.db "select prompt(prompt, 'markov') from prompts"

Now let's try a more complex one,

llm install llm-gpt4all

tsellm prompts.db "select prompt(prompt, 'orca-2-7b') from prompts"

Interactive Shell

You can enter an interactive REPL-style shell to manipulate your database.

tsellm prompts.db

There you can do queries the normal way:

ALTER TABLE prompts ADD COLUMN orca;
UPDATE PROMPTS SET orca=prompt(prompt, 'orca-2-7b');

Installation

pip install tsellm

How

tsellm relies on the following facts:

• SQLite is bundled with the standard Python library (import sqlite3)
• Python 3.12 ships with a SQLite interactive shell
• one can create Python-written user-defined functions to be used in SQLite queries (see create_function)
• Simon Willison has gone through the process of creating the beautiful llm Python library and CLI

Development

pip install -e '.[test]'
pytest