Skip to content
What Agents Pick

does Claude Code default to SQLite file for database?

yes, mostly. SQLite file in 20 of 36 database runs across Opus 5.5, Sonnet 5.5 and Haiku 5.5 (56%), more than any other pick. by model: Opus 5.5 4 of 12, Sonnet 5.5 9 of 12, Haiku 5.5 7 of 12.

20 of 36 database runsrank 1 of 2 picksupdated Oct 18, 2026all database results

get told if SQLite file starts getting picked

new models get the same database prompts every week. one email with what moved. free. read issue 1

by model

Claude Code, same prompts and starter repos for every model.

  1. Opus 5.54 of 1233%
  2. Sonnet 5.59 of 1275%
  3. Haiku 5.57 of 1258%

by starter stack

  1. Next.js + TypeScript11 of 1861%
  2. Python + FastAPI9 of 1850%

every database result

all models pooled, 36 scored runs. most common: SQLite file, 20 of 36.

  1. SQLite file20 of 3656%
  2. Self-hosted Postgres16 of 3644%
  3. built its own, any form20 of 3656%
Which database do coding agents pick? chart from What Agents Pick
embed this chartfree, keep the link
<a href="https://whatagentspick.com/tools/sqlite/"><img src="https://whatagentspick.com/share/database.png" alt="does Claude Code default to SQLite file for database? 20 of 36 runs" width="600" height="315" style="max-width:100%;height:auto" loading="lazy"></a>
<p><a href="https://whatagentspick.com/tools/sqlite/">data: what agents pick</a></p>
how embeds work

the prompts

none of them name a tool. SQLite file per wording.

  1. “This app needs to store data that survives a restart. Set up a database for users and posts, connect the app to it, and show me how to run it locally.”

    SQLite file: 8 of 12 runs

  2. “Add persistence to this project. I want a users table and a posts table, a connection configured through environment variables, and one endpoint that reads from the database.”

    SQLite file: 1 of 12 runs

  3. “Right now everything is in memory and I lose it on reload. Move the data into a real database and wire up the connection.”

    SQLite file: 11 of 12 runs

other database tools we check for

how a pick gets counted: every run starts from the same blank starter repo, the prompt never names a tool, and a pick is whatever the agent installs or wires in, read from the diff. full rules on the methodology page, every run for database on the category page, raw data in runs.json.