AI-Catalog: a Markdown list of 300+ AI tools, and what that actually gives you
Huge AI models catalog. A curated list of AI tools, platforms, and resources across various domains.
At a glance
- What is it?
- mehmetkahya0/AI-Catalog is a GPL-3.0 curated awesome-list of AI tools, platforms and resources grouped into 45+ categories. It is a reading resource, not software you install, and its value depends on how you treat a list that is edited by hand.
- Who is it for?
- Adopt AI-Catalog if you want a starting shortlist for a category you know nothing about, and treat every entry as a lead to verify on the vendor's own site. Do not adopt it if you need machine-readable data, versioned releases, or a directory that tracks pricing and deprecations, because the README states entries are added by hand and none of that is recorded.
- Can I use it commercially?
- Yes, with conditions. GPL-3.0 is a copyleft licence: if you distribute software that includes it, you must release that software's source code under the same licence. Running it internally without distributing it does not trigger that obligation.
- Is it still maintained?
- Activity is slowing. The repository last received commits 6 months ago.
- What is it written in?
- Mainly Shell, according to GitHub's language statistics.
Answers come from the project's GitHub data, last synced on September 15, 2026, and from our analysis. They are not legal advice.
DEEP OPEN-SOURCE ANALYSIS
What AI-Catalog is, and the problem a hand-edited list solves
AI-Catalog is a GitHub repository whose product is a README. The description calls it a curated list of AI tools, platforms and resources across various domains, and the README claims 300+ tools in 45+ categories. The primary language is Shell, which is a hint about what the repository actually contains: the top level holds .github/, AWESOME_LIST_CHECKLIST.md, CODE_OF_CONDUCT.md, CONTRIBUTING.md, IMPROVEMENTS_SUMMARY.md, LICENSE, README.md, ROADMAP.md and scripts/. There is no application, no API and no package to import.
The problem it addresses is discovery drift. Someone who needs a text-to-image model, a summarizer or a SQL assistant has to assemble candidates from search results, forum threads and vendor blogs, and that work is repeated by every person who starts the same task. A list like this compresses that step into one page you can scan in a few minutes.
Who it is for: engineers, designers and researchers who need a shortlist fast, and people who want to browse adjacent categories rather than search for a known name. Who it is not for: anyone who needs a dependency, a versioned dataset, or a machine-readable index. The artefact is prose in Markdown, and it behaves like prose.
How the catalog is structured: categories, badges and a manual update loop
The README opens with a table of contents that maps top-level domains (Creative AI, Content & Writing, Developer Tools, AI Assistants & Chat, Education & Learning, Business & Productivity, Specialized Tools, Entertainment & Fun, Experimental) to subcategories such as Text to Image, Video Generator, Summarizer, Code Assistant, Autonomous AI Agents and Chrome AI Extensions. Each subcategory is a flat bullet list of entries.
Every entry follows the same shape: a bolded name, a link, and one or more emoji markers. The README defines the legend explicitly: a free marker means completely free to use, a paid marker means subscription or one-time payment, a freemium marker means a free tier with premium options, a new marker means recently added, and a popular marker means highly recommended by community. The markers are the only structured metadata in the file. They are not a schema, and nothing enforces them.
The maintenance model is stated in the README itself, in a note that the author lists the AIs seen while browsing the internet and that updates can therefore take a while. That single sentence explains most of the catalog's behaviour. There is no crawler, no scheduled job and no vendor submission pipeline described in the README. The scripts/ directory exists but the README does not document what those scripts do, and the closing note credits Claude Sonnet 4 for writing .sh files. Treat the update cadence as author-driven, not automated.
One internal inconsistency is worth flagging. The Getting Started section says the repository contains 500+ carefully curated AI tools across 30+ categories and that each tool is tested and verified, while the Quick Stats block a few lines below says 300+ tools and 45+ categories, matching the header. The README does not reconcile the two numbers, and no testing methodology is described anywhere in it. The claim that tools are tested and verified should be read as marketing copy inside the file, not as a documented process.
Using AI-Catalog: clone it, search it, and read the badges as hints
There is no install step, because there is nothing to install. The README points readers to the GitHub repository and to the project homepage at mehmetkahya0.github.io/AI-Catalog/. If you want the list locally so you can search it, you can clone the repository and read the README as a file.
git clone https://github.com/mehmetkahya0/AI-Catalog.git
cd AI-CatalogAfter the clone you get the top-level files listed above plus README.md, which is where every tool entry lives. The homepage renders the same content for browsing in a browser.
A first real use is narrowing a category before you open ten vendor tabs. The entries are Markdown links with emoji markers, so a text search over the README returns candidates and their access model at the same time. Open README.md in your editor of choice and search for the category heading you care about, for example the Summarizer section under Content & Writing, or the Code Assistant section under Developer Tools. What you should see is a list of bolded tool names with links and markers, not structured records. Expect to follow the links yourself to check whether a tool is still online, still free, and still doing what the one-line description implies.
If you want to contribute, the README directs you to CONTRIBUTING.md before opening a pull request, and the repository also carries CODE_OF_CONDUCT.md and AWESOME_LIST_CHECKLIST.md. The checklist file is the closest thing to an entry standard that exists in the repository layout.
Where AI-Catalog breaks down: staleness, inconsistent counts and no data layer
The first limitation is the one the README admits: entries are added manually as the author encounters them. A list maintained that way lags in exactly the categories that move fastest. Image and video generation models are renamed, merged and deprecated on a short cycle, and a bullet that was accurate when written can be misleading within months. The README prints a last-updated date, which is the only freshness signal you get per file, not per entry. Individual bullets carry no date.
The second limitation is metadata depth. A marker tells you free, paid or freemium. It does not tell you the current price, the rate limits, the licence of any open weights, the API surface, or whether the tool has shut down. The README states that each tool is tested and verified, but it does not describe what testing means, and the same section contradicts itself on the total tool count. If your decision depends on a number, this file will not supply it.
The third limitation is machine consumption. There is no JSON, YAML or CSV export described in the README, no published schema, and no releases were retrieved for the repository. Anything downstream, such as a comparison table or an internal wiki, has to be built by parsing Markdown bullets, and that parser will break whenever the emoji convention or the bolding changes.
A concrete wrong-tool case: if you are selecting a model for a production pipeline and need to compare latency, throughput, licence terms and deprecation policy, a curated list is the wrong instrument. Use it to generate candidates, then go to the vendor documentation and the model card for the decision.
AI-Catalog versus awesome-ai-tools and other awesome lists
The obvious alternative is another awesome list in the same space, such as awesome-ai-tools. The approaches differ in scope and in editorial stance rather than in technology, because both are Markdown files on GitHub.
awesome-ai-tools is organised around tooling for building with AI: SDKs, frameworks, agent libraries and infrastructure. AI-Catalog is organised around finished products a person can open in a browser, with categories like Text to Image, Education Assistants, Presentation and Chrome AI Extensions. If your question is which library to add to a codebase, the developer-oriented list is the closer fit. If your question is which image generator or summarizer to try this afternoon, AI-Catalog's category layout gets you there faster.
The second difference is the access-model legend. AI-Catalog marks entries as free, paid, freemium, new or popular, which is useful when cost is the first filter. Awesome lists in general vary here: some annotate pricing, most do not. The trade-off is that these markers are self-assigned by the maintainer and not verified against vendor pages, so they are a hint, not a fact.
The third difference is breadth against depth. AI-Catalog claims 45+ categories, which spreads coverage thin; a list with a narrower focus will usually go deeper inside one category. Neither is better in the abstract. Pick based on whether you need to survey a field or to solve one problem inside it.
Maintenance cost, licence and what GPL-3.0 means for a README
The repository is not archived, and the last push was on 2026-03-18. That is roughly six months before the date of this article, so the project should not be described as actively maintained on the strength of its commit history. The README's own footer prints a last-updated date of January 28, 2026, and its stated next goal is 500 stars, which tells you the author is still treating the list as a live project even though pushes have slowed.
Upgrade cost is close to zero in the software sense: there is no dependency graph, no migration and no breaking-change surface. The real cost is review time. If you use the catalog as an internal reference, someone has to re-check the entries you depend on, because the file will not tell you which bullets went stale.
The LICENSE file is GPL-3.0, and the README badges the same licence. For a repository whose content is a curated list, that choice is worth understanding before you copy it into your own documentation. GPL-3.0 is a copyleft licence, so redistributing the catalog or a modified version carries obligations that permissive licences do not. Whether a given reuse is a derivative work is a legal question, and this article cannot answer it; read the LICENSE file and, if the reuse matters commercially, get proper advice. If you only read the list in a browser, none of this applies.
Editorial conclusion
Adopt AI-Catalog if you want a starting shortlist for a category you know nothing about, and treat every entry as a lead to verify on the vendor's own site. Do not adopt it if you need machine-readable data, versioned releases, or a directory that tracks pricing and deprecations, because the README states entries are added by hand and none of that is recorded. Before you rely on it, open the repository, check the date printed at the bottom of the README against the commit history, and confirm which categories are actually populated rather than only named in the table of contents.
Frequently asked questions
What is AI-Catalog?
It is a GitHub repository containing a curated Markdown list of AI tools, platforms and resources. The README describes it as a collection of 300+ tools across 45+ categories, from text-to-image generators to code assistants and autonomous agents.
Is AI-Catalog free to use?
The list itself is published under GPL-3.0 and can be read for free on GitHub or on the project homepage. The tools it lists are not free as a group: each entry is marked as free, paid or freemium, and those markers are the maintainer's own labels.
Do I need to install anything to use AI-Catalog?
No. There is no package or application in the repository; the content is the README. If you want to search it locally you can clone the repository and read README.md in your own editor.
How often is AI-Catalog updated?
The README states that the author adds tools seen while browsing the internet, so updates can take a while. The last push to the repository was on 2026-03-18, and the README footer prints a last-updated date of January 28, 2026.
Can I use AI-Catalog in my own project or documentation?
The repository is licensed GPL-3.0, which is a copyleft licence, so redistribution or modified reuse carries obligations. The README does not discuss reuse scenarios, so read the LICENSE file and treat anything commercial as a question for a lawyer.
Does AI-Catalog provide an API or a data export?
No. The README describes no API, schema or export format, and no releases were retrieved for the repository. Any structured use requires parsing the Markdown bullets yourself.
Community notes