Dwon

CUDA_SETUP.md

@@ -0,0 +1,176 @@
+# CUDA 12.8 and PyTorch Setup Guide
+
+This guide covers setting up LingBot-World with CUDA 12.8 and PyTorch.
+
+## Prerequisites
+
+- NVIDIA GPU with compute capability 7.0+ (e.g., RTX 2000 series or newer)
+- Docker with NVIDIA Container Toolkit installed
+- At least 16GB of system RAM
+- Sufficient disk space for models and data
+
+## Option 1: Docker Setup (Recommended)
+
+### 1. Install Docker and NVIDIA Container Toolkit
+
+**Ubuntu/Debian:**
+```bash
+# Install Docker
+curl -fsSL https://get.docker.com -o get-docker.sh
+sudo sh get-docker.sh
+
+# Install NVIDIA Container Toolkit
+distribution=$(. /etc/os-release;echo $ID$VERSION_ID)
+curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add -
+curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | sudo tee /etc/apt/sources.list.d/nvidia-docker.list
+
+sudo apt-get update
+sudo apt-get install -y nvidia-container-toolkit
+sudo systemctl restart docker
+```
+
+**Windows:**
+- Install Docker Desktop for Windows
+- Ensure WSL2 backend is enabled
+- Install NVIDIA drivers for WSL2
+
+### 2. Build and Run with Docker Compose
+
+```bash
+# Build the image
+docker-compose build
+
+# Run the container
+docker-compose up -d
+
+# Enter the container
+docker-compose exec lingbot-world bash
+
+# Or run directly
+docker-compose run --rm lingbot-world python your_script.py
+```
+
+### 3. Verify CUDA Installation
+
+Inside the container:
+```python
+import torch
+print(f"PyTorch version: {torch.__version__}")
+print(f"CUDA available: {torch.cuda.is_available()}")
+print(f"CUDA version: {torch.version.cuda}")
+print(f"Number of GPUs: {torch.cuda.device_count()}")
+if torch.cuda.is_available():
+    print(f"GPU Name: {torch.cuda.get_device_name(0)}")
+```
+
+## Option 2: Local Installation
+
+### 1. Install CUDA 12.8
+
+**Ubuntu/Debian:**
+```bash
+# Download and install CUDA 12.8
+wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.1-1_all.deb
+sudo dpkg -i cuda-keyring_1.1-1_all.deb
+sudo apt-get update
+sudo apt-get install cuda-toolkit-12-8
+
+# Add to PATH
+echo 'export PATH=/usr/local/cuda-12.8/bin:$PATH' >> ~/.bashrc
+echo 'export LD_LIBRARY_PATH=/usr/local/cuda-12.8/lib64:$LD_LIBRARY_PATH' >> ~/.bashrc
+source ~/.bashrc
+```
+
+**Windows:**
+1. Download CUDA 12.8 from [NVIDIA Developer](https://developer.nvidia.com/cuda-downloads)
+2. Run the installer and follow the prompts
+3. Verify installation: `nvcc --version`
+
+### 2. Create Python Environment
+
+```bash
+# Create virtual environment
+python3.10 -m venv venv
+source venv/bin/activate  # On Windows: venv\Scripts\activate
+
+# Upgrade pip
+pip install --upgrade pip setuptools wheel
+```
+
+### 3. Install PyTorch with CUDA 12.4 Support
+
+**Note:** PyTorch doesn't have official CUDA 12.8 builds yet. Using CUDA 12.4 builds which are compatible:
+
+```bash
+# Install PyTorch with CUDA 12.4 support
+pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu124
+```
+
+### 4. Install LingBot-World Dependencies
+
+```bash
+# Install remaining dependencies
+pip install -r requirements.txt
+
+# Install in editable mode
+pip install -e .
+```
+
+### 5. Verify Installation
+
+```bash
+python -c "import torch; print('CUDA available:', torch.cuda.is_available())"
+```
+
+## Troubleshooting
+
+### CUDA Version Mismatch
+
+If you encounter CUDA version mismatch errors:
+- PyTorch CUDA 12.4 builds are compatible with CUDA 12.x runtime
+- Ensure your NVIDIA driver supports CUDA 12.8 (driver version ≥ 550.x)
+
+### Out of Memory Errors
+
+If you encounter OOM errors:
+- Reduce batch size in your training scripts
+- Enable gradient checkpointing
+- Use mixed precision training (fp16 or bf16)
+
+### Flash Attention Installation Issues
+
+If `flash_attn` fails to install:
+```bash
+# Install with CUDA support
+pip install flash-attn --no-build-isolation
+```
+
+Or build from source:
+```bash
+git clone https://github.com/Dao-AILab/flash-attention.git
+cd flash-attention
+python setup.py install
+```
+
+## Performance Optimization
+
+### Enable TF32 for Ampere+ GPUs
+
+```python
+import torch
+torch.backends.cuda.matmul.allow_tf32 = True
+torch.backends.cudnn.allow_tf32 = True
+```
+
+### Use Compilation (PyTorch 2.x)
+
+```python
+model = torch.compile(model)
+```
+
+## Additional Resources
+
+- [NVIDIA CUDA Downloads](https://developer.nvidia.com/cuda-downloads)
+- [PyTorch Installation Guide](https://pytorch.org/get-started/locally/)
+- [NVIDIA Container Toolkit](https://github.com/NVIDIA/nvidia-docker)
+- [Flash Attention](https://github.com/Dao-AILab/flash-attention)

Dockerfile

@@ -0,0 +1,51 @@
+# LingBot-World Dockerfile with CUDA 12.8 support
+FROM nvidia/cuda:12.8.0-cudnn9-devel-ubuntu22.04
+
+# Set environment variables
+ENV DEBIAN_FRONTEND=noninteractive
+ENV PYTHONUNBUFFERED=1
+ENV CUDA_HOME=/usr/local/cuda
+ENV PATH=${CUDA_HOME}/bin:${PATH}
+ENV LD_LIBRARY_PATH=${CUDA_HOME}/lib64:${LD_LIBRARY_PATH}
+
+# Install system dependencies
+RUN apt-get update && apt-get install -y \
+    python3.10 \
+    python3-pip \
+    python3-dev \
+    git \
+    wget \
+    curl \
+    build-essential \
+    libgl1-mesa-glx \
+    libglib2.0-0 \
+    libsm6 \
+    libxext6 \
+    libxrender-dev \
+    libgomp1 \
+    ffmpeg \
+    && rm -rf /var/lib/apt/lists/*
+
+# Upgrade pip
+RUN python3 -m pip install --upgrade pip setuptools wheel
+
+# Set working directory
+WORKDIR /app
+
+# Copy requirements and install Python dependencies
+COPY requirements.txt .
+RUN pip install --no-cache-dir torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu124
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Install Hugging Face CLI and authenticate using the command line
+RUN huggingface-cli login --token ${HUGGINGFACE_TOKEN} && \
+    python3 -c "from huggingface_hub import hf_hub_download; hf_hub_download(repo_id='google/embeddinggemma-300m', filename='config.json', repo_type='model', cache_dir='/app/checkpoints')"
+
+# Copy project files
+COPY . .
+
+# Install the package in editable mode
+RUN pip install -e .
+
+# Set the default command
+CMD ["/bin/bash"]

README copy.md

@@ -0,0 +1,126 @@
+<div align="center">
+  <img src="assets/teaser.png">
+
+<h1>LingBot-World: Advancing Open-source World Models</h1>
+
+Robbyant Team
+
+</div>
+
+
+<div align="center">
+
+[![Page](https://img.shields.io/badge/%F0%9F%8C%90%20Project%20Page-Demo-00bfff)](https://technology.robbyant.com/lingbot-world)
+[![Tech Report](https://img.shields.io/badge/%F0%9F%93%84%20Tech%20Report-Document-teal)](LingBot_World_paper.pdf)
+[![Paper](https://img.shields.io/static/v1?label=Paper&message=PDF&color=red&logo=arxiv)](https://github.com/robbyant/lingbot-world)
+[![Model](https://img.shields.io/static/v1?label=%F0%9F%A4%97%20Model&message=HuggingFace&color=yellow)](https://huggingface.co/robbyant/lingbot-world-base-cam)
+[![Model](https://img.shields.io/static/v1?label=%F0%9F%A4%96%20Model&message=ModelScope&color=purple)](https://www.modelscope.cn/models/Robbyant/lingbot-world-base-cam)
+[![License](https://img.shields.io/badge/License-Apache--2.0-green)](LICENSE.txt)
+
+</div>
+
+-----
+
+We are excited to introduce **LingBot-World**, an open-sourced world simulator stemming from video generation. Positioned
+as a top-tier world model, LingBot-World offers the following features. 
+- **High-Fidelity & Diverse Environments**: It maintains high fidelity and robust dynamics in a broad spectrum of environments, including realism, scientific contexts, cartoon styles, and beyond. 
+- **Long-Term Memory & Consistency**: It enables a minute-level horizon while preserving contextual consistency over time, which is also known as long-term memory. 
+- **Real-Time Interactivity & Open Access**: It supports real-time interactivity, achieving a latency of under 1 second when producing 16 frames per second. We provide public access to the code and model in an effort to narrow the divide between open-source and closed-source technologies. We believe our release will empower the community with practical applications across areas like content creation, gaming, and robot learning.
+
+## 🎬 Video Demo
+<div align="center">
+  <video src="https://github.com/user-attachments/assets/ea4a7a8d-5d9e-4ccf-96e7-02f93797116e" width="100%" poster=""> </video>
+</div>
+
+## 🔥 News
+- Jan 29, 2026: 🎉 We release the technical report, code, and models for LingBot-World.
+
+<!-- ## 🔖 Introduction of LingBot-World
+We present **LingBot-World**, an **open-sourced** world simulator stemming from video generation. Positioned
+as a top-tier world model, LingBot-World offers the following features. 
+- It maintains high fidelity and robust dynamics in a broad spectrum of environments, including realism, scientific contexts, cartoon styles, and beyond. 
+- It enables a minute-level horizon while preserving contextual consistency over time, which is also known as **long-term memory**. 
+- It supports real-time interactivity, achieving a latency of under 1 second when producing 16 frames per second. We provide public access to the code and model in an effort to narrow the divide between open-source and closed-source technologies. We believe our release will empower the community with practical applications across areas like content creation, gaming, and robot learning. -->
+
+## ⚙️ Quick Start
+This codebase is built upon [Wan2.2](https://github.com/Wan-Video/Wan2.2). Please refer to their documentation for installation instructions.
+### Installation
+Clone the repo:
+```sh
+git clone https://github.com/robbyant/lingbot-world.git
+cd lingbot-world
+```
+Install dependencies:
+```sh
+# Ensure torch >= 2.4.0
+pip install -r requirements.txt
+```
+Install [`flash_attn`](https://github.com/Dao-AILab/flash-attention):
+```sh
+pip install flash-attn --no-build-isolation
+```
+### Model Download
+
+| Model | Control Signals | Resolution | Download Links |
+| :---  | :--- | :--- | :--- |
+| **LingBot-World-Base (Cam)** | Camera Poses | 480P & 720P | 🤗 [HuggingFace](https://huggingface.co/robbyant/lingbot-world-base-cam) 🤖 [ModelScope](https://www.modelscope.cn/models/Robbyant/lingbot-world-base-cam) |
+| **LingBot-World-Base (Act)** | Actions | - | *To be released* |
+| **LingBot-World-Fast**       |    -    | - | *To be released* |
+
+
+Download models using modelscope-cli:
+ ```sh
+pip install modelscope
+modelscope download robbyant/lingbot-world-base-cam --local_dir ./lingbot-world-base-cam
+```
+### Inference
+Our model supports video generation at both 480P and 720P resolutions. You can find data samples for inference in the `examples/` directory, which includes the corresponding input images, prompts, and control signals. To enable long video generation, we utilize multi-GPU inference powered by FSDP and DeepSpeed Ulysses.
+- 480P:
+
+This means the frame_num must be in the form of 4n + 1, where n is an integer (e.g., 1, 2, 3, etc.). For example, valid values include 5, 9, 13, 161, 321, etc.
+
+python  generate.py --task i2v-A14B --size 480*832 --ckpt_dir lingbot-world-base-cam --image examples/00/image.jpg --action_path examples/00 --frame_num 31 --prompt "The video presents a soaring journey through a fantasy jungle. The wind whips past the rider's blue hands gripping the reins, causing the leather straps to vibrate. The ancient gothic castle approaches steadily, its stone details becoming clearer against the backdrop of floating islands and distant waterfalls."  --save_file C:\workspace\world\lingbot-world\out
+
+ python  generate.py --task i2v-A14B --size 480*832 --ckpt_dir lingbot-world-base-cam --image examples/00/image.jpg --action_path examples/00 --frame_num 21 --prompt "The video presents a soaring journey through a fantasy jungle. The wind whips past the rider's blue hands gripping the reins, causing the leather straps to vibrate. The ancient gothic castle approaches steadily, its stone details becoming clearer against the backdrop of floating islands and distant waterfalls."  
+
+``` sh
+$env:USE_LIBUV=0
+torchrun --nproc_per_node=1 generate.py --task i2v-A14B --size 480*832 --ckpt_dir lingbot-world-base-cam --image examples/00/image.jpg --action_path examples/00 --dit_fsdp --t5_fsdp --ulysses_size 8 --frame_num 161 --prompt "The video presents a soaring journey through a fantasy jungle. The wind whips past the rider's blue hands gripping the reins, causing the leather straps to vibrate. The ancient gothic castle approaches steadily, its stone details becoming clearer against the backdrop of floating islands and distant waterfalls."
+```
+  python -m torch.distributed.run --nproc_per_node=1 generate.py --task i2v-A14B --size 480*832 --ckpt_dir lingbot-world-base-cam --image examples/00/image.jpg --action_path examples/00 --dit_fsdp --t5_fsdp --ulysses_size 8 --frame_num 161 --prompt "The video presents a soaring journey through a fantasy jungle. The wind whips past the rider's blue hands gripping the reins, causing the leather straps to vibrate. The ancient gothic castle approaches steadily, its stone details becoming clearer against the backdrop of floating islands and distant waterfalls.
+
+- 720P:
+``` sh
+torchrun --nproc_per_node=8 generate.py --task i2v-A14B --size 720*1280 --ckpt_dir lingbot-world-base-cam --image examples/00/image.jpg --action_path examples/00 --dit_fsdp --t5_fsdp --ulysses_size 8 --frame_num 161 --prompt "The video presents a soaring journey through a fantasy jungle. The wind whips past the rider's blue hands gripping the reins, causing the leather straps to vibrate. The ancient gothic castle approaches steadily, its stone details becoming clearer against the backdrop of floating islands and distant waterfalls."
+```
+Alternatively, you can run inference without control actions:
+``` sh
+torchrun --nproc_per_node=8 generate.py --task i2v-A14B --size 480*832 --ckpt_dir lingbot-world-base-cam --image examples/00/image.jpg --dit_fsdp --t5_fsdp --ulysses_size 8 --frame_num 161 --prompt "The video presents a soaring journey through a fantasy jungle. The wind whips past the rider's blue hands gripping the reins, causing the leather straps to vibrate. The ancient gothic castle approaches steadily, its stone details becoming clearer against the backdrop of floating islands and distant waterfalls."
+```
+Tips:
+If you have sufficient CUDA memory, you may increase the `frame_num` parameter to a value such as 961 to generate a one-minute video at 16 FPS.
+
+## 📚 Related Projects
+- [HoloCine](https://holo-cine.github.io/)
+- [Ditto](https://editto.net/)
+- [WorldCanvas](https://worldcanvas.github.io/)
+- [RewardForcing](https://reward-forcing.github.io/)
+- [CoDeF](https://qiuyu96.github.io/CoDeF/)
+
+## 📜 License
+This project is licensed under the Apache 2.0 License. Please refer to the [LICENSE file](LICENSE.txt) for the full text, including details on rights and restrictions.
+
+## ✨ Acknowledgement
+We would like to express our gratitude to the Wan Team for open-sourcing their code and models. Their contributions have been instrumental to the development of this project.
+
+## 📖 Citation
+If you find this work useful for your research, please cite our paper:
+
+```
+@article{lingbot-world,
+      title={Advancing Open-source World Models}, 
+      author={Robbyant Team},
+      journal={arXiv preprint arXiv:xx.xx},
+      year={2026}
+}
+```