Master

.gitignore

@@ -7,6 +7,7 @@ queue_images/
 modules/toolbox/model_esrgan/
 modules/toolbox/model_rife/
 .framepack/
+.claude/
 modules/toolbox/data/
 modules/toolbox/bin
 queue.json

INSTALLATION_NOTES.md

@@ -0,0 +1,139 @@
+# FramePack Studio Installation Notes
+
+## Installation Summary
+
+**Installation Method:** Fresh venv created and packages installed via automated process
+**Date:** December 29, 2025
+**Python Version:** 3.13
+
+### Installed Packages
+
+**Core AI/ML Packages:**
+- PyTorch: 2.8.0+cu128
+- TorchVision: 0.23.0+cu128
+- TorchAudio: 2.8.0+cu128
+- Diffusers: 0.36.0
+- Transformers: 4.57.3
+- Accelerate: 1.12.0
+- PEFT: 0.18.0
+
+**UI & Web:**
+- Gradio: 6.2.0
+
+**Media Processing:**
+- OpenCV Contrib Python: 4.12.0.88
+- av: 16.0.1
+- imageio: 2.37.2
+- imageio-ffmpeg: 0.6.0
+- decord: 0.6.0
+
+**Upscaling/Enhancement:**
+- facexlib: 0.3.0
+- gfpgan: 1.3.8
+- realesrgan: 0.3.0
+
+**Utilities:**
+- sentencepiece: 0.2.1
+- torchsde: 0.2.6
+- scipy: 1.16.3
+- numpy: 2.2.6
+
+### CUDA Configuration
+- CUDA Version: 12.8
+- Optimized for RTX 40xx and 50xx GPUs
+
+### Known Issues
+
+#### BasicSR Not Installed
+BasicSR has compatibility issues with Python 3.13 and could not be installed.
+
+**Impact:**
+- Some advanced toolbox features (ESRGAN, GFPGAN, RealESRGAN) may have limited functionality
+- Main FramePack Studio features should work normally
+
+**Workaround:**
+1. Use Python 3.10, 3.11, or 3.12 if BasicSR features are critical
+2. Wait for BasicSR to release Python 3.13 compatible wheels
+
+**Alternative:** You can try installing BasicSR manually later if needed:
+```bash
+venv\Scripts\activate
+pip install basicsr --no-build-isolation
+```
+
+#### NumPy Version
+- Requirements specify numpy==1.26.2 but numpy 2.2.6 was installed
+- This is a newer version with better Python 3.13 support
+- Most packages are compatible with NumPy 2.x
+
+### Optional Acceleration Packages
+
+The install script offers optional acceleration packages:
+
+**Sage Attention:**
+- Requires: triton-windows<3.4
+- Significant speed improvements for RTX 40xx/50xx
+- Pre-built wheels available for Python 3.10-3.12
+
+**Flash Attention:**
+- Alternative acceleration method
+- Pre-built wheels available
+
+**Note:** These may not be available for Python 3.13 yet.
+
+### Scripts Created
+
+1. **install_40xx_50xx.bat** - Full installation script for RTX 40xx/50xx GPUs
+2. **start.bat** - Launches FramePack Studio with venv activation
+3. **activate_venv.bat** - Activates venv for manual commands
+
+### Running the Application
+
+```bash
+# Option 1: Use the start script
+start.bat
+
+# Option 2: Use the existing run script
+run.bat
+
+# Option 3: Manual
+venv\Scripts\activate
+python studio.py
+```
+
+### Verifying Installation
+
+```bash
+venv\Scripts\activate
+python -c "import torch; print(f'PyTorch: {torch.__version__}'); print(f'CUDA: {torch.cuda.is_available()}')"
+```
+
+Expected output:
+```
+PyTorch: 2.8.0+cu128
+CUDA: True
+```
+
+## Troubleshooting
+
+### If PyTorch CUDA is not detected:
+1. Ensure NVIDIA drivers are up to date
+2. Verify nvidia-smi shows your GPU
+3. Reinstall PyTorch with CUDA 12.8
+
+### If packages are missing:
+```bash
+venv\Scripts\activate
+pip install -r requirements.txt
+```
+
+### Clean Reinstall:
+1. Delete the `venv` folder
+2. Run `install_40xx_50xx.bat` again
+
+## Python Version Recommendation
+
+**Recommended:** Python 3.10, 3.11, or 3.12
+**Current:** Python 3.13 (newer, some packages may have limited compatibility)
+
+If you experience issues, consider using Python 3.12 for maximum compatibility.

activate_venv.bat

@@ -0,0 +1,24 @@
+@echo off
+echo ============================================
+echo Activating FramePack-Studio Virtual Environment
+echo ============================================
+echo.
+
+REM Check if venv exists
+if not exist "%cd%\venv\Scripts\activate.bat" (
+    echo Error: Virtual environment not found!
+    echo Please run install_40xx_50xx.bat first to set up the environment.
+    echo.
+    pause
+    exit /b 1
+)
+
+echo Virtual environment activated.
+echo You can now run Python commands within this environment.
+echo.
+echo To deactivate, type: deactivate
+echo To run FramePack-Studio, type: python studio.py
+echo.
+
+REM Activate and keep command prompt open
+cmd /k "%cd%\venv\Scripts\activate.bat"

diffusers_helper/models/hunyuan_video_packed.py

@@ -63,7 +63,7 @@
 has_xformers = xformers_attn_func is not None
 
 if has_sage:
-    print("✅  Using SAGE Attention (highest performance).")
+    print("[OK] Using SAGE Attention (highest performance).")
     ignored = []
     if has_flash:
         ignored.append("Flash Attention")
@@ -72,16 +72,16 @@
     if ignored:
         print(f"   - Ignoring other installed attention libraries: {', '.join(ignored)}")
 elif has_flash:
-    print("✅  Using Flash Attention (high performance).")
+    print("[OK] Using Flash Attention (high performance).")
     if has_xformers:
         print("   - Consider installing SAGE Attention for highest performance.")
         print("   - Ignoring other installed attention library: xFormers")
 elif has_xformers:
-    print("✅  Using xFormers.")
+    print("[OK] Using xFormers.")
     print("   - Consider installing SAGE Attention for highest performance.")
     print("   - or Consider installing Flash Attention for high performance.")
 else:
-    print("⚠️  No attention library found. Using native PyTorch Scaled Dot Product Attention.")
+    print("[WARNING] No attention library found. Using native PyTorch Scaled Dot Product Attention.")
     print("   - For better performance, consider installing one of:")
     print("     SAGE Attention (highest performance), Flash Attention (high performance), or xFormers.")
 print("-------------------------------\n")