Update README.md

sammrai · Jan 10, 2025 · dd8a845 · dd8a845
1 parent 51fa641
commit dd8a845
Showing 1 changed file with 28 additions and 42 deletions.
diff --git a/README.md b/README.md
@@ -1,35 +1,26 @@
 # Stable Diffusion WebUI Forge Docker Image
 
-This Docker image is pre-configured for [Stable Diffusion WebUI Forge](https://github.com/lllyasviel/stable-diffusion-webui-forge) ** (not Auto1111 WebUI), offering a streamlined way to run the WebUI with all necessary components bundled. By using this Docker image, you gain access to the following features:
+This Docker image is pre-configured for [Stable Diffusion WebUI Forge](https://github.com/lllyasviel/stable-diffusion-webui-forge) (not Auto1111 WebUI), offering a streamlined way to run the WebUI with all necessary components bundled. By using this Docker image, you gain access to the following features:
 
 - **Pre-configured WebUI Forge setup**
 - **Optional CivitAI model downloader integration**
 - **Optional Cloudflare Tunnel configuration**
 
 Simply clone the repository and start the service with `docker compose up -d`. Note that models are not included and must be downloaded separately using the provided options.
 
----
 
 ## Supported CUDA Versions
 
-To ensure compatibility between your host machine's NVIDIA driver version and the appropriate Docker image, please refer to the following table:
+Ensure that the NVIDIA driver version on your **host machine** is compatible with the selected Docker image. Refer to the table below for details:
 
-| Docker Image                                 | Compatible Host NVIDIA Driver Version (Linux) | Forge Configuration                                      |
-|----------------------------------------------|-----------------------------------------------|----------------------------------------------------------|
-| `sammrai/sd-forge-docker:12.4.0`             | 550.54.14 or higher                           | CUDA 12.4 + Pytorch 2.4 (Fastest, but MSVC may be broken, xformers may not work)   |
-| `sammrai/sd-forge-docker:12.1.0`             | 530.30.02 or higher                           | CUDA 12.1 + Pytorch 2.3.1 (Recommended)                                              |
+| Docker Image                                 | Required NVIDIA Driver Version (Host Machine)  | Forge Configuration                                      |
+|----------------------------------------------|------------------------------------------------|----------------------------------------------------------|
+| `sammrai/sd-forge-docker:12.4.0`             | 550.54.14 or higher                            | CUDA 12.4 + Pytorch 2.4 (Fastest, but MSVC issues possible, xformers may not work)   |
+| `sammrai/sd-forge-docker:12.1.0`             | 530.30.02 or higher                            | CUDA 12.1 + Pytorch 2.3.1 (Recommended)                                              |
 
-**Important:** Ensure that your host machine's NVIDIA driver version falls within the specified range for the chosen Docker image.
-**Note:** The driver version references above are based on publicly available information and do not guarantee actual compatibility or official endorsement.
-
-You can check your current driver version by running:
-
-```bash
-nvidia-smi
-```
-
-This command will display the `Driver Version` installed on your system.
-For more detailed information on CUDA and driver compatibility, please refer to NVIDIA's official documentation. 
+**Key Notes:**
+1. The NVIDIA driver version **on the host machine** must meet or exceed the required version for the selected Docker image.
+2. The listed driver versions are based on public documentation and do not guarantee full compatibility or official endorsement.
 
 ## Setup Instructions
 
@@ -52,7 +43,6 @@ docker compose up -d
 
 **Note:** Models are not included by default. You can either download them manually or use the optional CivitAI integration.
 
----
 
 ## Deployment Options
 
@@ -110,36 +100,32 @@ For secure external access to the WebUI, configure a **Cloudflare Tunnel**. Foll
 
    Refer to the [Cloudflare Tunnel Documentation](https://developers.cloudflare.com/cloudflare-one/connections/connect-apps/) for detailed setup instructions. This includes authenticating with Cloudflare and establishing the tunnel using your `TUNNEL_TOKEN`.
 
----
 
 ## CivitAI Integration (Optional)
 
 The **CivitAI** model downloader automates the downloading and placement of models. While optional, it is **highly recommended** to add your **CIVITAI\_TOKEN** in the `.env` file, as many models require authentication to download.
 
-### 1. Add Your CIVITAI Token
-
-Include your **CIVITAI\_TOKEN** in the `.env` file to enable seamless integration:
-
-```env
-CIVITAI_TOKEN=yourtoken...
-```
+1. **Add Your CIVITAI Token**  
+   Include your **CIVITAI\_TOKEN** in the `.env` file to enable seamless integration:
 
-### 2. Supported Model Types and Example Commands
-
-The following model types are supported for automatic downloads, along with their corresponding aliases to ensure models are placed in the correct directories:
+   ```env
+   CIVITAI_TOKEN=yourtoken...
+   ```
 
-- **Lora**: `@lora`
-- **VAE**: `@vae`
-- **Embed**: `@embed`
-- **Checkpoint**: `@checkpoint`
+2. **Supported Model Types and Example Commands**  
+   The following model types are supported for automatic downloads, along with their corresponding aliases to ensure models are placed in the correct directories:
+
+   - **Lora**: `@lora`
+   - **VAE**: `@vae`
+   - **Embed**: `@embed`
+   - **Checkpoint**: `@checkpoint`
 
-Here are some example commands to download and place models:
-
-```bash
-docker compose exec webui civitdl 257749 439889 @checkpoint
-docker compose exec webui civitdl 332646 @embed
-docker compose exec webui civitdl 660673 @vae
-docker compose exec webui civitdl 341353 @lora
-```
+   Here are some example commands to download and place models:
 
+   ```bash
+   docker compose exec webui civitdl 257749 439889 @checkpoint
+   docker compose exec webui civitdl 332646 @embed
+   docker compose exec webui civitdl 660673 @vae
+   docker compose exec webui civitdl 341353 @lora
+   ```