VMamba: Visual State Space Models，code is based on mamba

Simba

LongLLaVA: Scaling Multi-modal LLMs to 1000 Images Efficiently via Hybrid Architecture

CLIP-Mamba: CLIP Pretrained Mamba Models with OOD and Hessian Evaluation

Self-training LLaVA for medical

A Large Language-Vision Assistant for Pathology Image Understanding (BIBM-2024)

Official Repository of Personalized Visual Instruct Tuning

[NeurIPS'23 Oral] Visual Instruction Tuning (LLaVA) built towards GPT-4V level capabilities and beyond.