docs: update README

README.md

@@ -14,6 +14,7 @@ tags:
 [![arXiv](https://img.shields.io/badge/arXiv-2410.08631-B31B1B?logo=arxiv&logoColor=white)](https://arxiv.org/abs/2410.08631)
 [![GitHub](https://img.shields.io/badge/GitHub-cryofm-181717?logo=github&logoColor=white)](https://github.com/ByteDance-Seed/cryofm)
 [![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
+[![Docs](https://img.shields.io/badge/Docs-cryofm-4CAF50?logo=read-the-docs&logoColor=white)](https://bytedance-seed.github.io/cryofm/docs/)
 
 </div>
 
@@ -60,6 +61,36 @@ The model is available in two variants optimized for different resolution requir
 
 ## Quick Start
 
+### Installation
+
+Before using CryoFM1, ensure you have:
+
+#### 1. Install CryoFM with compatible dependencies
+
+CryoFM1 uses the HDiT model architecture, which depends on the `natten` package. Different versions of `natten` have varying requirements for PyTorch and CUDA versions. For a reproducible installation, follow these steps:
+
+```bash
+# natten 0.17.5 uses type union syntax, you must use python >=3.10
+conda create -n cryofm python=3.10 -y
+conda activate cryofm
+
+# Install PyTorch 2.5.1 with CUDA 12.4 support
+pip install torch==2.5.1 torchvision==0.20.1 --index-url https://download.pytorch.org/whl/cu124
+
+# Install natten 0.17.5 compatible with PyTorch 2.5.0 and CUDA 12.4
+pip install natten==0.17.5+torch250cu124 -f https://whl.natten.org
+
+# Clone and install CryoFM
+git clone https://github.com/ByteDance-Seed/cryofm
+cd cryofm
+pip install .
+```
+
+#### 2. Download model checkpoints and configuration files
+
+Download the CryoFM1 model weights and configuration files from the [Hugging Face repository](https://huggingface.co/ByteDance-Seed/cryofm-v1).
+
+
 ### Unconditional Generation
 
 CryoFM1 provides two model variants for different resolution needs:
@@ -132,7 +163,12 @@ for i in range(3):
     )
 ```
 
-### Ethical Considerations
+### Downstream Tasks
+
+CryoFM1 demonstrates various downstream tasks including density map denoising, anisotropy noise correction, and missing wedge restoration. For detailed instructions on how to run these tasks, please refer to the [Downstream Tasks documentation](https://bytedance-seed.github.io/cryofm/docs/model-guides/cryofm1/downstream-tasks.html).
+
+
+## Ethical Considerations
 
 This model is intended for scientific research and structural biology applications. Users should:
 - Ensure proper attribution when using generated structures