🔥🔥 LLaVA++: Extending LLaVA with Phi-3 and LLaMA-3 (LLaVA LLaMA-3, LLaVA Phi-3)

Official code for Paper "Mantis: Multi-Image Instruction Tuning" [TMLR 2024]

A light llama-like llm inference framework based on the triton kernel.

[ICCVW 25] LLaVA-MORE: A Comparative Study of LLMs and Visual Backbones for Enhanced Visual Instruction Tuning

A Large Language-Vision Assistant for Pathology Image Understanding (BIBM-2024 & Journal of Artificial Intelligence Review 2025)

This repository contains a web application designed to execute relatively compact, locally-operated Large Language Models (LLMs).

An implementation of FastVLM/LLaVA or any llm/vlm model using FastAPI (backend) and react js (backend) + Action/Caption mode and frame control

OllamaMulti-RAG 🚀 is a multimodal AI chat app combining Whisper AI for audio, LLaVA for images, and Chroma DB for PDFs, enhanced with Ollama and OpenAI API. 📄 Built for AI enthusiasts, it welcomes contributions—features, bug fixes, or optimizations—to advance practical multimodal AI research and development collaboratively.

A useful collection of notebooks for quantization, fine-tuning, and inference with the Turkish LLaVA model.

A Python CLI tool that uses a local LLAMA model to describe media.

A command-line tool for evaluating student artwork submissions with written explanations. This tool analyzes images containing both visual work and written components, providing feedback and scoring.

Generate an alternate text for an image locally on your machine.

With the integration of LLAVA, Mistral, LangChain, and Streamlit, the chatbot developed has shown notable advancement in its functionality. The chatbot “Converso” offers multi-modal capabilities which includes basic text-based chatting, image-based interactions and audio based chatting

ScreenSage as an AI testing assistant that generates test cases of based on screenshots of the features of your newly developed app.