CLIP: Connecting Text and Images
OpenAI published its CLIP milestone on January 5, 2021, introducing Contrastive Language–Image Pre-training as a neural network that learns visual concepts from natural-language supervision and can be applied to visual classification tasks by specifying category names in natural language.
Historical classification
This verified record belongs to LINEAiGE's artificial-intelligence history layer beginning in 1956.
Why it matters
A documented multimodal-learning milestone connecting natural-language supervision with transferable visual recognition and zero-shot evaluation.
People, organizations & technologies
PEOPLE
Alec Radford, Ilya Sutskever, Jong Wook Kim, Gretchen Krueger, Sandhini Agarwal
ORGANIZATIONS
OpenAI
TECHNOLOGIES
CLIP, Contrastive Language–Image Pre-training
Evidence & verification
STATE
verified
CONFIDENCE
high
LAST VERIFIED
2026-09-08T20:49:08+08:00
The January 5 2021 milestone date, CLIP name, natural-language-supervision description, zero-shot visual-classification framing, and listed OpenAI milestone authors are supported directly by OpenAI's first-party CLIP page. The date is attached only to that documented public milestone; no internal training-completion date, geography, ancestry, influence, causality, or relationships are inferred. The later arXiv submission date is retained as corroborating paper metadata rather than substituted for the earlier first-party public milestone.
Sources
CLIP: Connecting text and images
OpenAI
PUBLISHED
2021-01-05
Learning Transferable Visual Models From Natural Language Supervision
arXiv
PUBLISHED
2021-02-26
Trace the chronology
Record identity
LINEAiGE IDclip-2021