AI responses, an independent measurement
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.
58%
37%
33%
4/12 use cases where AIs disagree
430 responses actually analyzed
#
The panelThe 14 compared tools View panel
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%.
| # | Tool | Average recommendation rate | |
|---|---|---|---|
| 1 | PostgreSQL | 58% | |
| 2 | MySQL | 37% | |
| 3 | SQLite | 33% | |
| 4 | MongoDB | 33% | |
| 5 | Supabase | 33% | |
| 6 | CockroachDB | 27% | |
| 7 | Redis | 24% | |
| 8 | DynamoDB | 22% | |
| 9 | Firebase Firestore | 12% | |
| 10 | ClickHouse | 12% | |
| 11 | DuckDB | 3% | |
| 12 | MariaDB | 2% | |
| 13 | Microsoft SQL Server | 1% | |
| 14 | Oracle Database | 1% |
## The tool each model spontaneously surfaces [¶](#par-modele)
Use case by use case, the tool each model most often puts first (every use case counts the same, chatty or not).
Tool most often highlighted by each assistant:
PostgreSQL (5) · CockroachDB (1)
For the same use case, a different dominant tool depending on the model ¶ #
## Which use cases AIs associate with which tool [¶](#terrains)
For each cell, the percentage shows the share of responses recommending the tool. Select a dot for details. ‘n’ is the number of responses analyzed; results based on few responses are indicative.
↔ Swipe the matrix to explore use cases.
| Tool | Database for a SaaS startup MVP | Database for a side project | First database for a beginner learning backend development | Database for a mobile app backend | Database for analytics and reporting | Database for a high-traffic app at scale | Database for an enterprise app with strict compliance | Database for an e-commerce store | Database for a real-time collaborative app | Database for an AI app with vector search (RAG) | Embedded or local-first database for a desktop app | Database for a serverless or edge deployment |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ClickHouse | ||||||||||||
| CockroachDB | ||||||||||||
| DuckDB | ||||||||||||
| DynamoDB | ||||||||||||
| Firebase Firestore | ||||||||||||
| MariaDB | ||||||||||||
| Microsoft SQL Server | ||||||||||||
| MongoDB | ||||||||||||
| MySQL | ||||||||||||
| Oracle Database | ||||||||||||
| PostgreSQL | ||||||||||||
| Redis | ||||||||||||
| SQLite | ||||||||||||
| Supabase |
Click a dot to show the detail of an association.
Your tool ranks poorly, or is missing? Submit it for the next wave →
[Are you a brand? See what AI says about you →](/en/mesurer)
## Why do AIs recommend each tool? [¶](#pourquoi)
The main argument the models invoke for each tool, the price they quote, and the caveat they attach.
×7
×3
×38
×8
×7
×4
×2
×25
×5
×5
×3
×15
×2
×16
×9
×7
×7
×18
×11
×8
×6
×21
×19
×12
×11
×13
×7
×4
×4
×49
×4
×2
×2
×19
×8
×6
×3
Which assistants give a clear recommendation? ¶ #
Share of answers where the model commits to one tool, vs “it depends…”, vs doesn't commit.
n=7063 37
n=7254 46
n=7253 42 6
n=7251 47 1
n=7239 58 3
n=7239 61
## The same need, a different profile, a different tool [¶](#profils)
For each buyer profile, the tools the models surface most (existing answers sliced by profile).
n=30
- ClickHouse97%
- PostgreSQL33%
- MySQL27%
alternative
n=24
- PostgreSQL21%
- Redis17%
- MySQL12%
alternative
n=24
- Supabase96%
- PostgreSQL46%
- Firebase Firestore42%
alternative
n=23
- MySQL87%
- PostgreSQL83%
- MongoDB70%
alternative
n=18
- PostgreSQL83%
- CockroachDB67%
- MongoDB44%
alternative
n=18
- Redis89%
- DynamoDB78%
- CockroachDB61%
alternative
Other tools that appeared spontaneously ¶ #
Comparable tools the models cited spontaneously, beyond the tracked panel: worth watching (or adding to the comparison).
How do AIs present each tool? ¶ #
Breakdown of favorable / neutral / critical verified mentions. Click a tool for a favorable and a critical verbatim.
How it is measured, and what it is not ¶ #
What we measure
*· claude-sonnet-5*
*· deepseek-v4-pro*
*· gemini-3.5-flash*
*· mistral-medium-3-5*
*· gpt-5.2-chat*
· grok-4.3
430 responses · 6 models · 12 use cases · 14 tools · query cost $5.31
Each question asks for advice without offering a list of tools. We measure the tools the assistant chooses to mention spontaneously, then how it presents them.
What these results do not prove
- Not a ranking of the tools' actual quality.
- Not a feature or pricing comparison.
- We measure model behavior, not the truth. Low-n rates are indicative.
To cite this page
“According to the Opinion Radar observatory dated July 20, 2026, PostgreSQL has the highest average recommendation rate in the tested situations (58%), ahead of MySQL (37%).”
Reference link: this page (public methodology included). Attribute the claims to the AI models, never to the tools themselves.
Ask the question yourself
You can reuse these questions to check the result. AI responses may nevertheless vary between attempts and change over time.
Your tool isn't measured, or ranks poorly?
Submit it: it joins the queue for the next wave and will be measured under the same conditions as the panel.
Discover what AIs recommend, and why
Are you a brand or a vendor? Measure precisely where AIs mention you (or don't) and why.