Flamingo Research Introduction
DeepMind introduced Flamingo, a visual language model for few-shot learning across multimodal tasks, in an official research post on April 28, 2022.
Historical classification
This verified record belongs to LINEAiGE's artificial-intelligence history layer beginning in 1956.
Why it matters
A documented multimodal research milestone in which a single visual language model was presented as handling interleaved images, videos, and text and adapting to multiple tasks from only a small number of examples.
People, organizations & technologies
ORGANIZATIONS
Google DeepMind
TECHNOLOGIES
Flamingo
Evidence & verification
STATE
verified
CONFIDENCE
high
LAST VERIFIED
2026-09-09T23:02:15+08:00
Google DeepMind's first-party research post is dated April 28, 2022 and explicitly says it introduces Flamingo as a single visual language model for few-shot learning over open-ended multimodal tasks. The associated arXiv preprint was submitted April 29, 2022 and describes Flamingo as a family of visual language models that can ingest interleaved visual and textual data. This record intentionally uses the dated first-party research introduction as the event boundary; it does not treat that date as a product launch, API release, unrestricted public-weight release, deployment date, or evidence of causal influence on later systems. No geography, ancestry, influence, causality, or relationships are inferred.
Sources
Tackling multiple tasks with a single visual language model
Google DeepMind
PUBLISHED
2022-04-28
Flamingo: a Visual Language Model for Few-Shot Learning
arXiv
PUBLISHED
2022-04-29
Trace the chronology
Record identity
LINEAiGE IDflamingo-2022