CC12M (Conceptual 12M)
by Google · free · Last verified 2026-03-17
CC12M is a large-scale dataset by Google containing 12 million image-text pairs from the web. It was created with a less restrictive filtering process than its predecessor, CC3M, to achieve greater scale and diversity. This makes it a foundational resource for pretraining large vision-language models like CLIP and ALIGN.
https://huggingface.co/datasets/laion/conceptual-captions-12m-webdataset ↗B
B—Above Average
Adoption: B+Quality: B+Freshness: CCitations: AEngagement: F
Specifications
- License
- Custom
- Pricing
- free
- Capabilities
- vision-language-pretraining, zero-shot-image-classification, text-to-image-retrieval, image-to-text-retrieval, visual-question-answering-pretraining, image-captioning-model-training, visual-concept-learning, benchmarking-foundation-models
- Integrations
- [object Object], [object Object], [object Object], [object Object]
- Use Cases
- [object Object], [object Object], [object Object], [object Object]
- API Available
- Yes
- Tags
- multimodal, image-text, web-crawl, vision-language, pretraining, large-scale-dataset, google-research, alt-text, noisy-data, foundation-models, zero-shot-learning
- Added
- 2026-03-17
- Completeness
- 0.9%
Index Score
65.1Adoption
73
Quality
77
Freshness
43
Citations
82
Engagement
0