fastvlm

Flask app to interact with FastVLM using CPU only

Desktop VLM: Real-time FastVLM analysis of video & textures with live compute shaders

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

Real-time video captioning powered by FastVLM

This description captures the essence of your project as an intelligent document processing platform that bridges OCR technology with interactive document annotation capabilities.

针对视障人群设计的基于Rokid Glasses的实时氛围感知系统

情绪识别&音乐疗愈Demo

Real-time on-device camera intelligence — a tiered perception cascade (detection + tracking → gated VLM → event-sourced scene state) that turns any live video stream into live-synced understanding.

Fork of Apple's ml-fastvlm with mlx-swift-lm 3.x compatibility fixes

Deploy FastVLM1.5B on RK3588 and run on it

Colaboratory上でFastVLM(apple/ml-fastvlm)をお試しするサンプル

React Native mobile application for running Vision Language Models (VLMs) locally on your device.

This repository contains the MLX-optimized implementation of "FastVLM: Efficient Vision Encoding for Vision Language Models" (CVPR 2025), fully updated to support Swift 6.2+ and the latest MLX framework. 📌 Note: Please check out the dev branch!

Real-time hazard detection and scene description for the visual impaired navigating outdoors. It's an on-device mobile app paired with Apple watch via Bluetooth connection featuring hands-free haptic and voice feedback .