MANZANO: A Simple and Scalable Unified Multimodal Model with a Hybrid Vision Tokenizer
Paper
•
2509.16197
•
Published
•
54
Collection
-
-
Qwen/Qwen3-Omni-30B-A3B-Instruct
Any-to-Any • 35B • Updated • 272k • 726 -
facebook/dinov3-vitb16-pretrain-lvd1689m
Image Feature Extraction • 85.7M • Updated • 363k • 82 -
nvidia/NV-Embed-v2
Feature Extraction • 8B • Updated • 207k • 482 -
zju-community/matchanything_eloftr
16.1M • Updated • 5.15k • 79 -
DINOv3
🦖19Similarity, Classification
-
DINOv3 Web/Sat Interactive Similarity
🦖17Visualize image patch similarity like in DINOv3 presentation
-
Sheets
🗂615Create and enrich datasets with AI
-
FastVLM: Efficient Vision Encoding for Vision Language Models
Paper • 2412.13303 • Published • 72 -
FastVLM WebGPU
🍎401Real-time video captioning powered by FastVLM
-
ByteDance/lynx
Image-to-Video • Updated • • 134 -
Talk2DINO
💻18Demo of Talk2DINO, model presented at ICCV 2025.
-
facebook/vjepa2-vitl-fpc64-256
Video Classification • 0.3B • Updated • 94.1k • 162 -
Yoloe
🚀63Detect and segment objects in images using text, visual, or prompt-free prompts
-
Distill Any Depth
💻172Generate depth maps from images
-
Jupyter Agent 2
🏃248Run code and analyze data in a Jupyter notebook
-
Describe Anything
⚡338Describe masked parts of images
-
amd/Nitro-E
Text-to-Image • Updated • 5.02k • 86 -
dx8152/Qwen-Edit-2509-Multiple-angles
Image-to-Image • Updated • 86.3k • • 737