Multitask Unified Model (MUM)

Published

May 18, 2021

Multitask Unified Model (MUM) is an AI model developed by Google (United States), first published in May 2021. It works in the language, multimodal and vision domain, on tasks such as language modeling/generation, image captioning, question answering and 4 more.

Epoch AI has no training-compute estimate for this model.

Access: Unreleased. Its weights are not openly released. It is built on top of T5-11B.

Full record

Organization: Google
Country of organization: United States
Domain: Language, Multimodal, Vision
Task: Language modeling/generation, Image captioning, Question answering, Visual question answering, Translation, Character recognition (OCR), Search
Model accessibility: Unreleased
Open weights: No
Base model: T5-11B