Skip to content

LLaVA

University of Wisconsin MadisonMicrosoft ResearchColumbia UniversityChatQuestion answeringVisual question answeringOpen weights

LLaVA is a chat model from University of Wisconsin Madison,Microsoft Research,Columbia University released in 2023 with 13000000000.0 parameters.

About LLaVA

Instruction tuning large language models (LLMs) using machine-generated instruction-following data has improved zero-shot capabilities on new tasks, but the idea is less explored in the multimodal field. In this paper, we present the first attempt to

Details

Provider
University of Wisconsin Madison,Microsoft Research,Columbia University
Task
Chat,Question answering,Visual question answering
Parameters
13000000000.0
Released
2023-04-17
Open weights
Yes
View model source

Explore

FAQ