Data · CC BY 4.0 · rebuilt nightly
The whole catalog, as data
257 AI tools, 45 use cases and every evidence score, in one JSON file that is rewritten every night. Free for any use, including commercial, if you credit AI Use Cases with a link.
Endpoints
no key, no account, CORS open| URL | Format | What it is |
|---|---|---|
| /data.json | JSON | Use cases, tools and the score for every tool. The file this site's own search runs on. |
| /llms.txt | Markdown | The directory written for language models: top 3 per use case with scores and URLs. |
| /feed.xml | RSS | New AI tool releases as the recorder sees them, hourly. |
| /sitemap.xml | XML | Every page, including the per-use-case free and open-source lists. |
Shape of data.json
{
"use_cases": [
{
"slug": "...",
"name": "...",
"question": "...",
"category": "...",
"keywords": [
"..."
]
}
],
"tools": [
{
"slug": "ollama",
"name": "Ollama",
"tagline": "...",
"use_cases": [
"run-llms-locally"
],
"plus": "...",
"minus": "...",
"pricing": "open-source",
"url": "https://ollama.com",
"logo": "ollama.png"
}
],
"rank": {
"ollama": {
"score": 99.0
}
},
"at": "2026-10-18T03:15:00"
}
Scores are 0-100 and only comparable within a rebuild — the normalisation is over the catalog on the night it ran. The formula is published, and every input is visible on each tool's page.
Licence
CC BY 4.0. Use it, sell it, train on it — attribute it as: "AI tool rankings from AI Use Cases (https://aiusecases.dev), 2026, CC BY 4.0" with a link.
What the data is not: an endorsement, a quality judgement, or a measure of how good a tool is at its job. It measures public evidence of use and maintenance, which is the only thing that can be checked from outside.