Vision1.8B official

moondream

by vikhyat

A tiny, fast vision-language model that runs almost anywhere — captioning and visual Q&A on the edge.

arcflare run moondream

[ 01 ] // quickstart

┌─ arcflare — zsh // tty1

[ 02 ] // readme

A small, efficient vision-language model designed to run on modest hardware, great for image captioning and visual question answering.

arcflare run moondream

[ 03 ] // relatedmore →

Hugging Face

Lightweight multimodal model that processes video, image and text inputs to generate text; about 5.2GB of GPU RAM for video inference.

17% of a 24 GB card

#vision #small

Google

Gemma 3: 1B to 27B, with image understanding from 4B up, 128K context and 140+ languages.

38% of a 24 GB card

#vision #chat #multilingual

Mistral AI

Mistral Small 3.1 24B: fast general model with image understanding and 128K context, under Apache-2.0.

67% of a 24 GB card

#vision #chat #tools