Which database do AI assistants recommend · Opinion Radar
AI responses, an independent measurement
Which database do AI assistants recommend
What 6 AIs recommend in this category
PostgreSQL leads overall, but the winner changes with the need.
Across the 12 situations tested, PostgreSQL is the tool most often recommended by the 6 assistants. But in 4 of 12 situations, the tool highlighted varies by AI. The choice therefore depends as much on the stated need as on the assistant used.
Measured on July 20, 2026 · 430 responses analyzed · 6 AI assistants surveyed
PO<br>PostgreSQL
58%
MY<br>MySQL
37%
SQ<br>SQLite
33%
4/12 use cases where AIs disagree<br>430 responses actually analyzed
Get the results of the next measurement : be notified when the answers change
Alert me
The panelThe 14 compared tools<br>View panel ＋
CLClickHouse
COCockroachDB
DUDuckDB
DYDynamoDB
FFFirebase Firestore
MAMariaDB
MSMicrosoft SQL Server
MOMongoDB
MYMySQL
ODOracle Database
POPostgreSQL
RERedis
SQSQLite
SUSupabase
⓪ The big picture
Which tools do AIs recommend most often? ¶
The average recommendation rate is calculated across all tested situations, weighted equally. One assistant may recommend several tools, so these percentages are not expected to add up to 100%.
#ToolAverage recommendation rate
POPostgreSQL
58%
MYMySQL
37%
SQSQLite
33%
MOMongoDB
33%
SUSupabase
33%
COCockroachDB
27%
RERedis
24%
DYDynamoDB
22%
FFFirebase Firestore
12%
10<br>CLClickHouse
12%
11<br>DUDuckDB
3%
12<br>MAMariaDB
2%
13<br>MSMicrosoft SQL Server
1%
14<br>ODOracle Database
1%
① The answer depends on which AI you use
The tool each model spontaneously surfaces ¶
Use case by use case, the tool each model most often puts first (every use case counts the same, chatty or not).
claude-sonnet-5<br>COCockroachDBleading on 2/12 use cases
deepseek-v4-pro<br>POPostgreSQLleading on 8/12 use cases
gemini-3.5-flash<br>POPostgreSQLleading on 8/12 use cases
mistral-medium-3-5<br>POPostgreSQLleading on 7/12 use cases
gpt-5.2-chat<br>POPostgreSQLleading on 5/12 use cases
grok-4.3<br>POPostgreSQLleading on 5/12 use cases
Tool most often highlighted by each assistant:<br>PostgreSQL (5) · CockroachDB (1)
② The models don't agree
For the same use case, a different dominant tool depending on the model ¶
Database for a SaaS startup MVP
consensus
Database for a side project
split
First database for a beginner learning backend development
consensus
Database for a mobile app backend
split
Database for analytics and reporting
consensus
Database for a high-traffic app at scale
consensus
Database for an enterprise app with strict compliance
consensus
Database for an e-commerce store
consensus
Database for a real-time collaborative app
split
Database for an AI app with vector search (RAG)
split
Embedded or local-first database for a desktop app
consensus
Database for a serverless or edge deployment
consensus
Database for a SaaS startup MVP<br>consensus
PO<br>PostgreSQL<br>4 models
“Solid runner-up: PostgreSQL (self-hosted first, later managed) If you know you’ll need strict transactional integrity across many concurrent writers, or if your team already feels at home with Postgres, go with vanilla PostgreSQL.”
deepseek-v4-pro · gemini-3.5-flash · gpt-5.2-chat · grok-4.3
SU<br>Supabase<br>2 models
“Supabase (Postgres + Realtime)”
claude-sonnet-5 · mistral-medium-3-5
Database for a side project<br>split
PO<br>PostgreSQL<br>3 models
“Supabase · PostgreSQL · Gives you a full‑powered Postgres with auto‑generated REST/GraphQL APIs, a nice web dashboard, and a free tier.”
deepseek-v4-pro · mistral-medium-3-5 · grok-4.3
SU<br>Supabase<br>2 models
“Supabase (The Best Overall for SaaS)”
gemini-3.5-flash · gpt-5.2-chat
SQ<br>SQLite<br>1 model
“Plain SQLite is genuinely underrated for this scale.”
claude-sonnet-5
First database for a beginner learning backend development<br>consensus
PO<br>PostgreSQL<br>6 models
“For a Django + React bootcamp, PostgreSQL is the best choice.”
claude-sonnet-5 · deepseek-v4-pro · gemini-3.5-flash · mistral-medium-3-5 · gpt-5.2-chat · grok-4.3
Database for a mobile app backend<br>split
DY<br>DynamoDB<br>2 models
“Where Document Stores (MongoDB, DynamoDB, Firestore) Win”
claude-sonnet-5 · deepseek-v4-pro
PO<br>PostgreSQL<br>2 models
“Relational Store (e.g., PostgreSQL, Aurora)”
gemini-3.5-flash · mistral-medium-3-5
FF<br>Firebase Firestore<br>1 model
“Firebase Firestore (or Realtime Database) – Strongest all-rounder for indies”
grok-4.3
RE<br>Redis<br>1 model
“✅ In-memory store (Redis)”
gpt-5.2-chat
Database for analytics and reporting<br>consensus
CL<br>ClickHouse<br>6 models
“For millions of events/day with sub-second aggregations, columnar storage (ClickHouse, Druid, BigQuery, Redshift, Snowflake) is almost certainly your answer.”
claude-sonnet-5 · deepseek-v4-pro · gemini-3.5-flash · mistral-medium-3-5 · gpt-5.2-chat · grok-4.3
Database for a high-traffic app at scale<br>consensus
CO<br>CockroachDB<br>5 models
“CockroachDB — Postgres wire-compatible, distributed writes,...