Stars
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
An open-source long-horizon SuperAgent harness that researches, codes, and creates. With the help of sandboxes, memories, tools, skill, subagents and message gateway, it handles different levels of…
🎬 500+ curated Seedance 2.0 video generation prompts — cinematic, anime, UGC, ads, meme styles. Includes Seedance API guides, character consistency tips, and advanced video workflows.
Academic Research Skills for Claude Code: research → write → review → revise → finalize
Comprehensive production pipeline for quad-modal AI filmmaking with Seedance 2.0
Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞
Unified automatic quality assessment for speech, music, and sound.
Scalable Minecraft multiplayer data collection engine
PyTorch code and models for VJEPA2 self-supervised learning from video.
An official implementation of DanceGRPO: Unleashing GRPO on Visual Generation
Official code for paper Advantage Weighted Matching: Aligning RL with Pretraining in Diffusion Models
Fast, Sharp & Reliable Agentic Intelligence
A framework for efficient model inference with omni-modality models
VeOmni: Scaling Any Modality Model Training with Model-Centric Distributed Recipe Zoo
NVIDIA FastGen: Fast Generation from Diffusion Models
Towards Scalable Pre-training of Visual Tokenizers for Generation
Qwen3-TTS is an open-source series of TTS models developed by the Qwen team at Alibaba Cloud, supporting stable, expressive, and streaming speech generation, free-form voice design, and vivid voice…
[ICLR 26 Oral] Stable Video Infinity: Infinite-Length Video Generation with Error Recycling
The official implementation of StereoPilot
(arXiv) MixGRPO: Unlocking Flow-based GRPO Efficiency with Mixed ODE-SDE
Edit-R1: Reinforce Image Editing with Diffusion Negative-Aware Finetuning and MLLM Implicit Feedback
[ICCV'25]DimensionX: Create Any 3D and 4D Scenes from a Single Image with Controllable Video Diffusion
HY-World 1.5: A Systematic Framework for Interactive World Modeling with Real-Time Latency and Geometric Consistency
[NeurIPS 2025] WorldMem: Long-term Consistent World Simulation with Memory
Accelerating MoE with IO and Tile-aware Optimizations
Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models
Official Python inference and LoRA trainer package for the LTX-2 audio–video generative model.