# Lightweight Multimodal

HyperCLOVAX-SEED-Vision-Instruct-3B is a lightweight multimodal model developed by NAVER, featuring image-text understanding and text generation capabilities, with special optimization for Korean language processing.

A multimodal model trained based on google/gemma-3-4b-it, specializing in high-quality data processing for mathematics, programming, science, and puzzle-solving domains.

A Japanese visual language model trained based on the NVILA-Lite architecture, supporting image-text interaction in both Japanese and English

This is a video-text-to-text model converted based on the MLX framework, suitable for video understanding and instruction-following tasks.

A lightweight multimodal model designed for analyzing video content, capable of processing video, image, and text inputs to generate text outputs.

SmolVLM2-256M-Video is a lightweight multimodal model specifically designed for analyzing video content, capable of processing video, image, and text inputs to generate text outputs.

T-lite-it-1.0 is an image-to-text model supporting Russian and English, converted from the GGUF format.

nanoLLaVA-1.5 is a vision-language model with under 1 billion parameters, designed specifically for edge devices—compact yet powerful.

Imp-v1.5-4B-Phi3 is a high-performance lightweight multimodal large model with only 4 billion parameters, built on the Phi-3 framework and SigLIP visual encoder.

moondream2 is a compact vision-language model specifically designed for efficient operation on edge devices, offering convenient deployment through the llamafile format.

nanoLLaVA is a 1B-parameter vision-language model specifically designed for edge devices, featuring efficient operation.

MiniCPM-V is an efficient lightweight multimodal model optimized for edge device deployment, supporting bilingual (Chinese-English) interaction and outperforming models of similar scale.

A 1.6B-parameter multimodal model combining SigLIP and Phi-1.5 architectures, supporting image understanding and Q&A tasks

TinyLLaVA is a compact large-scale multimodal model framework focused on vision-language tasks, featuring small parameter size yet excellent performance.

UForm-Gen-Chat is the fine-tuned multimodal conversational version of UForm-Gen, primarily used for image caption generation and visual question answering tasks.

UForm-Gen is a small generative vision-language model primarily used for image caption generation and visual question answering.