GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization Paper • 2601.05242 • Published Jan 8 • 224
Mind-Brush: Integrating Agentic Cognitive Search and Reasoning into Image Generation Paper • 2602.01756 • Published 10 days ago • 22
3D-Aware Implicit Motion Control for View-Adaptive Human Video Generation Paper • 2602.03796 • Published 8 days ago • 55
PaperBanana: Automating Academic Illustration for AI Scientists Paper • 2601.23265 • Published 12 days ago • 170
DynamicVLA: A Vision-Language-Action Model for Dynamic Object Manipulation Paper • 2601.22153 • Published 13 days ago • 68
Running 107 The Eiffel Tower Llama 📝 107 Explore the Eiffel Tower Llama experiment with open-source models
Running on Zero MCP Featured 1.7k Z Image Turbo 🏃 1.7k Generate custom images from text prompts with size options
Running Featured 103 Supertonic TTS WebGPU ⚡ 103 Blazingly fast text-to-speech 100% locally in your browser
view article Article Transformers v5: Simple model definitions powering the AI ecosystem +2 Dec 1, 2025 • 296
Kandinsky 5.0: A Family of Foundation Models for Image and Video Generation Paper • 2511.14993 • Published Nov 19, 2025 • 231
view article Article Introducing smolagents: simple agents that write actions in code. +1 Dec 31, 2024 • 1.17k
view reply To understand clearly, you upload the Perquet DS (I do need to store it somewhere, and Perquet is optimized on Hub) here on the Hub and use the streaming feature while having a constant net connection, right?