Datasets:

tiagozip
/

tenor

Name: tenor
Creator: tiago
License: https://choosealicense.com/licenses/other/

Modalities:

Formats:

Size:

Tags:

Libraries:

License:

You need to agree to share your contact information to access this dataset

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

by downloading this dataset, you agree with the dataset license

metadata for about 15 million Tenor GIFs, crawled after Google shut new Tenor API keys down on 2026-06-30, updated monthly.

from datasets import load_dataset

gifs = load_dataset("tiagozip/tenor", "gifs", split="train", streaming=True)

config	rows	notes
gifs	15M+	one file per month the GIF was created, `data/gifs/YYYY-MM.parquet`
packs	13.8k	one row per media pack

column	type	notes
`id`	uint64	the number at the end of a `tenor.com/view/...` URL
`title`	string	Tenor's display title
`description`	string	Tenor's generated alt text
`tags`	list[string]	search tags
`username`	string	uploader
`userid`	string	null when Tenor did not expose one
`partnername`	string	content partner, null for user uploads
`content_rating`	string	`G`, `PG`, `PG-13` or `R`
`hasaudio`	bool
`created`	datetime	upload time, UTC
`itemurl`	string	`tenor.com/view/...` link
`width`	uint16
`height`	uint16
`duration`	float32	seconds, `0` for non-animated entries
`shares`	uint64	Tenor's share counter at crawl time
`pack`	string	which pack holds this GIF's MP4
`pack_offset`	uint64	byte offset of the MP4 inside that pack
`pack_length`	uint32	byte length of the MP4

no media right now. the packs the pack columns point into are not hosted anywhere at the moment.
not complete. Tenor has no sitemap, so this was built by crawling search. it is broad, not exhaustive.
shares is a snapshot taken while crawling.

by downloading the data, you agree to the license terms. tl;dr:

no republishing, sharing, or selling the data (linking here is fine)
no using it, or publishing anything made with it, without clear and obvious credit
no machine learning, AI training, or publishing embeddings and similar representations
research and publication are allowed, with credit and approval
you must honour opt-outs and deletions

the license covers this metadata collection. the GIFs themselves belong to whoever made them.

if something here is yours and you want it gone, email hi@tiago.zip with the id and it is removed from the next update.