Spaces:
Sleeping
Sleeping
docs: Update design and readme to reflect ComfyUI integration and .env config
Browse files
README.md
CHANGED
|
@@ -5,7 +5,8 @@ RPGPortrait is a Gradio-based web application that helps users build highly deta
|
|
| 5 |
## Features
|
| 6 |
- **25+ Character Parameters**: Deep customization including Identity, Appearance, Equipment, Environment, VFX, and Technical settings.
|
| 7 |
- **🧠 Gemini Refinement**: Use Gemini 3 Pro to intelligently refine your technical prompts into vivid, artistic descriptions.
|
| 8 |
-
- **🖼️
|
|
|
|
| 9 |
- **💾 Save & Load**: Export your character configurations as JSON files and import them back to restore your exact selections.
|
| 10 |
- **⚔️ Exotic Arsenal**: Expanded collection of exotic weapons, armor, and accessories (Katanas, Bone Armor, Arcane Plate, etc.).
|
| 11 |
- **🎒 Dual Accessories**: Select up to two different accessories for your character.
|
|
@@ -29,7 +30,14 @@ RPGPortrait is a Gradio-based web application that helps users build highly deta
|
|
| 29 |
```
|
| 30 |
5. **Set up Gemini API**:
|
| 31 |
- Create a `.env` file in the root directory.
|
| 32 |
-
- Add your
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
|
| 34 |
## Usage
|
| 35 |
|
|
|
|
| 5 |
## Features
|
| 6 |
- **25+ Character Parameters**: Deep customization including Identity, Appearance, Equipment, Environment, VFX, and Technical settings.
|
| 7 |
- **🧠 Gemini Refinement**: Use Gemini 3 Pro to intelligently refine your technical prompts into vivid, artistic descriptions.
|
| 8 |
+
- **🖼️ Multi-Backend Image Gen**: Toggle between **Gemini (Cloud)** and **ComfyUI (Local)** for image synthesis.
|
| 9 |
+
- **Workflow Injection**: Automated prompt, resolution, and seed injection into custom ComfyUI workflows.
|
| 10 |
- **💾 Save & Load**: Export your character configurations as JSON files and import them back to restore your exact selections.
|
| 11 |
- **⚔️ Exotic Arsenal**: Expanded collection of exotic weapons, armor, and accessories (Katanas, Bone Armor, Arcane Plate, etc.).
|
| 12 |
- **🎒 Dual Accessories**: Select up to two different accessories for your character.
|
|
|
|
| 30 |
```
|
| 31 |
5. **Set up Gemini API**:
|
| 32 |
- Create a `.env` file in the root directory.
|
| 33 |
+
- Add your keys and connection info:
|
| 34 |
+
```env
|
| 35 |
+
GEMINI_API_KEY=your_api_key_here
|
| 36 |
+
COMFY_HOST=127.0.0.1
|
| 37 |
+
COMFY_PORT=8188
|
| 38 |
+
OLLAMA_HOST=127.0.0.1
|
| 39 |
+
OLLAMA_PORT=11434
|
| 40 |
+
```
|
| 41 |
|
| 42 |
## Usage
|
| 43 |
|
design.md
CHANGED
|
@@ -17,7 +17,9 @@ The UI is built using `gradio.Blocks` to allow for a structured, multi-column la
|
|
| 17 |
- **Generation Suite**:
|
| 18 |
- Large technical prompt textbox (read-only).
|
| 19 |
- **🧠 Refine with Gemini**: Calls Gemini 3 Pro to create a vivid prompt.
|
| 20 |
-
- **🖼️ Generate Image**:
|
|
|
|
|
|
|
| 21 |
- **Output Area**:
|
| 22 |
- Image display for the portrait.
|
| 23 |
- 📥 Download button (PNG) for the generated image.
|
|
@@ -26,7 +28,9 @@ The UI is built using `gradio.Blocks` to allow for a structured, multi-column la
|
|
| 26 |
The app uses a 3-stage prompt pipeline:
|
| 27 |
1. **Technical Segmenting**: Combines dropdown values and extra info text into a base technical prompt using a YAML-based template.
|
| 28 |
2. **AI Refinement (Gemini)**: The technical prompt is sent to `gemini-3-pro-preview` with specialized system instructions to "art-ify" the description.
|
| 29 |
-
3. **Image Synthesis
|
|
|
|
|
|
|
| 30 |
|
| 31 |
## Data & Persistence
|
| 32 |
- **YAML Configuration**: `features.yaml` stores all possible dropdown values, their descriptive labels, and the final prompt template.
|
|
@@ -38,4 +42,6 @@ The app uses a 3-stage prompt pipeline:
|
|
| 38 |
- **UI Framework**: Gradio
|
| 39 |
- **AI Models**:
|
| 40 |
- **Text**: `gemini-3-pro-preview`
|
| 41 |
-
|
|
|
|
|
|
|
|
|
| 17 |
- **Generation Suite**:
|
| 18 |
- Large technical prompt textbox (read-only).
|
| 19 |
- **🧠 Refine with Gemini**: Calls Gemini 3 Pro to create a vivid prompt.
|
| 20 |
+
- **🖼️ Generate Image**: Routes to selected backend (Gemini or ComfyUI).
|
| 21 |
+
- **Backend Selector**:
|
| 22 |
+
- Radio buttons to toggle between **Gemini (Cloud)** and **ComfyUI (Local)**.
|
| 23 |
- **Output Area**:
|
| 24 |
- Image display for the portrait.
|
| 25 |
- 📥 Download button (PNG) for the generated image.
|
|
|
|
| 28 |
The app uses a 3-stage prompt pipeline:
|
| 29 |
1. **Technical Segmenting**: Combines dropdown values and extra info text into a base technical prompt using a YAML-based template.
|
| 30 |
2. **AI Refinement (Gemini)**: The technical prompt is sent to `gemini-3-pro-preview` with specialized system instructions to "art-ify" the description.
|
| 31 |
+
3. **Image Synthesis**:
|
| 32 |
+
- **Cloud**: Sent to `imagen-4.0-generate-001`.
|
| 33 |
+
- **Local**: Sent to a ComfyUI endpoint via `POST /prompt`, utilizing a custom workflow (`comfy_rpg_char_gen.json`).
|
| 34 |
|
| 35 |
## Data & Persistence
|
| 36 |
- **YAML Configuration**: `features.yaml` stores all possible dropdown values, their descriptive labels, and the final prompt template.
|
|
|
|
| 42 |
- **UI Framework**: Gradio
|
| 43 |
- **AI Models**:
|
| 44 |
- **Text**: `gemini-3-pro-preview`
|
| 45 |
+
- **Image Cloud**: `imagen-4.0-generate-001`
|
| 46 |
+
- **Image Local**: ComfyUI (Local Server)
|
| 47 |
+
- **Configuration**: `.env` for API keys and connection hosts/ports (ComfyUI, Ollama).
|