kakaobrain/coyo-dataset resource

Health v2 · maintenance only

Flags: no_releases archived no_license

Adoption not part of the score

1255 stars · 38 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

COYO-700M is a large-scale dataset of 747M image-text pairs scraped from CommonCrawl HTML documents, with extensive image- and text-level filtering and meta-attributes. The repository provides documentation, download tooling, and subset releases like COYO-Labeled-300M for training vision-language foundation models.

Use cases

• train text-to-image diffusion models
• pretrain CLIP-style contrastive vision-language models
• get large-scale image caption pairs for multimodal research
• download a web-crawled image-text dataset as an alternative to LAION
• filter and prepare web-scraped training data
• train large-scale vision-language foundation models

When to avoid

• you need curated, human-annotated image captions
• you only need a small clean dataset for fine-tuning
• you require a permissive license - the repo has no license file

Facets

dataset · maturity maintenance

machine-learning data-science etl machine-learning computer-vision artificial-intelligence python cross-platform image-text-pairs multimodal vision-language common-crawl foundation-models text-to-image alt-text natural-language-processing

Member repositories

For agents

markdown · JSON · MCP: product_card(name="kakaobrain/coyo-dataset")