Datasets:

tarekmasryo
/

genai-tools-platforms-data

pretty_name: Generative AI Tools & Platforms (2025)
license: cc-by-4.0
language:
  - en
size_categories:
  - 100<n<1K
tags:
  - generative-ai
  - llm
  - tools
  - platforms
  - catalog
  - benchmarking
  - ecosystem
  - eda
  - tabular
  - dataset
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/genai_tools_platforms_2025.csv

🧠 Generative AI Tools & Platforms (2025)

Author: Tarek Masryo · Kaggle
License: CC BY 4.0 (Attribution)

A clean, structured catalog of 113 Generative AI tools and platforms released up to 2025.

Each row represents one tool/platform and includes:

Vendor/maintainer
Canonical category and primary modality
API availability and status
Open-source indicator
Release year + derived timeline features
Modality flags and breadth count

Designed for ecosystem mapping, benchmarking, and analytics workflows.

📦 What’s inside

Canonical file: data/genai_tools_platforms_2025.csv
Unit of analysis: 1 row = 1 tool/platform
Rows: 113
Columns: 22

Column groups

Core info: tool_name, company, website, source_domain
Taxonomy: category_canonical, modality_canonical
Access: open_source, api_available, api_status
Timeline: release_year, years_since_release
Modality flags: mod_text, mod_image, mod_video, mod_audio, mod_code, mod_design, mod_infra, mod_productivity, mod_safety, mod_multimodal
Derived: modality_count (sum of supported modality flags)

📚 Data dictionary

Column	Type	Description
tool_name	string	Tool/platform name (intended unique key)
company	string	Vendor/maintainer
website	string	Official homepage (full URL)
source_domain	string	Extracted domain for grouping (derived from `website`)
category_canonical	string	Standardized use-case family (e.g., LLMs, Image Gen, Video Gen, RAG)
modality_canonical	string	Primary capability type (e.g., text, image, video, audio, code, multimodal)
open_source	int (0/1)	1 if marked open-source, else 0
api_available	int (0/1)	1 if a public API is available, else 0
api_status	string	API status (`api` or `unavailable`)
release_year	int	First public release/launch year
years_since_release	int	Years since release (computed as `2025 - release_year`)
mod_text	int (0/1)	Supports text modality
mod_image	int (0/1)	Supports image modality
mod_video	int (0/1)	Supports video modality
mod_audio	int (0/1)	Supports audio modality
mod_code	int (0/1)	Supports code modality
mod_design	int (0/1)	Supports design modality
mod_infra	int (0/1)	Supports infra/dev tooling modality
mod_productivity	int (0/1)	Supports productivity modality
mod_safety	int (0/1)	Supports safety/moderation modality
mod_multimodal	int (0/1)	Marked as multimodal (as provided in the dataset)
modality_count	int	Sum of supported modality flags

🛠️ How to use

Load with Datasets (recommended)

from datasets import load_dataset, get_dataset_config_names

repo_id = "tarekmasryo/genai-tools-platforms-data"

print("Configs:", get_dataset_config_names(repo_id))

ds = load_dataset(repo_id)  # loads the default config
df = ds["train"].to_pandas()

print(df.shape)
print(df.head(3))

Quick checks

# API coverage rate
api_rate = df["api_available"].mean()
print("API availability:", round(float(api_rate), 2))

# Category distribution
print(df["category_canonical"].value_counts().head(10))

# Modality breadth
print(df["modality_count"].describe())

🧭 Suggested use cases

Ecosystem maps and dashboards for the GenAI landscape
Benchmark API and open-source coverage across categories
Trend views by release year and timeline analytics
Cluster tools by capability breadth and modality mix
Simple recommendation baselines (rule-based or ML-based)

📝 Notes

This is a curated catalog compiled from public sources (official websites, docs, product pages).
Coverage is best-effort and may not be exhaustive.
Fields like open_source and api_available reflect publicly advertised status at collection time.

🧾 Citation

Tarek Masryo. (2025). Generative AI Tools & Platforms (2025).

📜 License

CC BY 4.0 (Attribution) — free to share and adapt with attribution.