Upload 3 files

DEPLOY.md

@@ -0,0 +1,70 @@
+# Deploying to HuggingFace Spaces
+
+## Quick Deploy
+
+1. **Create a new Space on HuggingFace**:
+   - Go to https://huggingface.co/spaces
+   - Click "Create new Space"
+   - Name: `PromptBridge-Demo` (or your choice)
+   - License: Apache 2.0
+   - SDK: Gradio
+   - Hardware: CPU (basic) - should be sufficient for 0.6B model
+
+2. **Upload files**:
+   ```bash
+   # Clone your new space
+   git clone https://huggingface.co/spaces/YOUR_USERNAME/YOUR_SPACE_NAME
+   cd YOUR_SPACE_NAME
+   
+   # Copy files from promptbridge-space/
+   cp ../promptbridge-space/app.py .
+   cp ../promptbridge-space/requirements.txt .
+   cp ../promptbridge-space/README.md .
+   
+   # Commit and push
+   git add .
+   git commit -m "Initial commit"
+   git push
+   ```
+
+3. **Alternative: Use HF CLI**:
+   ```bash
+   # From the langchain directory
+   huggingface-cli upload YOUR_USERNAME/YOUR_SPACE_NAME promptbridge-space --repo-type=space
+   ```
+
+## Local Testing
+
+Test locally before deploying:
+```bash
+cd promptbridge-space
+python app.py
+```
+
+Then visit http://127.0.0.1:7860
+
+## Hardware Requirements
+
+- **CPU Basic**: Works well for this 0.6B model
+- **CPU Upgrade**: Slightly faster inference
+- **GPU**: Overkill for this small model
+
+## Configuration Options
+
+Edit `app.py` to customize:
+- Default temperature (currently 0.7)
+- Default max_tokens (currently 512)
+- Examples
+- UI colors/theme
+
+## Tips
+
+1. **First load is slow**: Model downloads from HF Hub (~1.2GB)
+2. **Subsequent loads are fast**: Model is cached
+3. **CPU inference**: Expect 1-3 seconds per generation
+4. **Consider caching examples**: Set `cache_examples=True` in Examples for faster demo
+
+## Monitoring
+
+Check Space logs at:
+`https://huggingface.co/spaces/YOUR_USERNAME/YOUR_SPACE_NAME/logs`

README.md

@@ -1,14 +1,37 @@
 ---
-title: PromptBridge Demo
-emoji: ⚡
-colorFrom: blue
-colorTo: indigo
+title: PromptBridge-0.6b-Alpha
+emoji: 🌉
+colorFrom: purple
+colorTo: blue
 sdk: gradio
-sdk_version: 6.3.0
+sdk_version: 4.44.0
 app_file: app.py
 pinned: false
 license: apache-2.0
-short_description: Expand keywords into full t2i prompts, or do the revese.
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# PromptBridge-0.6b-Alpha Demo
+
+Interactive demo for [PromptBridge-0.6b-Alpha](https://huggingface.co/retowyss/PromptBridge-0.6b-Alpha), a specialized model for bidirectional prompt transformation for text-to-image generation.
+
+## Features
+
+- **Prompt Expansion**: Convert brief keywords into detailed image generation prompts
+- **Prompt Compression**: Distill lengthy descriptions into sentences or keywords
+- **Variation Generation**: Chain operations for creative prompt variations
+
+## Model Info
+
+- **Base**: Qwen3-0.6B
+- **Training**: 300K synthetic prompt pairs
+- **Domain**: Single adult humanoid subjects
+- **Content**: PG to R-rated (X-rated removed)
+
+## Usage
+
+1. Select a mode (Expand/Compress to Sentence/Compress to Keywords)
+2. Enter your prompt
+3. Adjust temperature and max tokens as needed
+4. Click Generate
+
+⚠️ **Note**: This model is specialized for prompt transformation only and cannot perform general tasks.

requirements.txt

@@ -0,0 +1,4 @@
+gradio==4.44.0
+transformers==4.46.0
+torch==2.5.1
+accelerate==1.1.1