ComfyUI-FlashVSR_Stable v.1.2 #69
Description
High-performance Video Super Resolution for ComfyUI with VRAM optimization.
Run FlashVSR on 8GB-24GB+ GPUs without artifacts. Features intelligent resource management, 5 VAE options, and auto-downloading models.
[1.2.0] - 2025-12-23
✨ New Features
5 New VAE Models: Added support for Wan2.2, LightVAE_W2.1, TAE_W2.2, LightTAE_HY1.5 (plus original Wan2.1).
Smart Resource Calculator: Added "Pre-flight" check to analyze VRAM/RAM usage and recommend optimal settings before processing.
Auto-Download: Added automatic downloading for missing VAE models from HuggingFace (lightx2v/Autoencoders repository).
Unified UI: Replaced complex vae_type and alt_vae inputs with a single vae_model dropdown.
🐛 Bug Fixes
Fixed Black Borders: Implemented correct "Padding → Process → Crop" logic to handle VAE dimension requirements without corrupting the output.
Fixed Video Glitches: Corrected Tensor Permutation logic (B,C,F,H,W vs B,H,W,C) to prevent visual artifacts.
Fixed Model Loading: Solved the issue where selecting "Wan2.2" incorrectly loaded "Wan2.1_VAE.pth".
Fixed OOM False Positive: OOM recovery no longer triggers prematurely (adjusted threshold to 95%).
⚡ Optimization
Unified Pipeline: Merged tiny, full, and tiny-long logic into a single optimized architecture.
OOM Protection: Improved memory management with 95% VRAM threshold using torch.cuda.mem_get_info().
Lossless Resize: Uses NEAREST interpolation for integer scaling (0.5, 0.25) to avoid blur.
Aggressive Garbage Collection: Added torch.cuda.empty_cache() before/after heavy VAE operations.
🛠 Refactoring
Explicit VAE Instantiation: No more state_dict inspection/guessing - strict class mapping based on user selection.
VAE_MODEL_MAP: Centralized configuration for all 5 VAE options (class, file, URL, dimensions).
Summary Logging: End-of-processing report with total time, peak VRAM, and resolutions.
Debug Logging: Shows selected_model vs loaded_model_path for verification.
[1.1.0] - 2025-12-22
🚀 New Features
Wan2.2 VAE Support: Integrated Wan2.2 VAE with optimized normalization statistics.
LightX2V VAE Integration: Added LightX2V VAE for ~50% VRAM reduction and 2-3x faster inference.
VAE Type Selection: Added vae_model dropdown in both Init Pipeline and Ultra-Fast nodes.
Factory Function: Added create_video_vae() for programmatic VAE selection.
⚡ Performance
VRAM Reduction: LightX2V reduces peak VRAM usage by approximately 50%.
Speed Improvement: LightX2V provides 2-3x faster VAE decode times.
🛠 Refactoring
Backward Compatibility: All new VAE types maintain full compatibility with existing Wan2.1 weights.
Architecture Constants: Added VAE_FULL_DIM, VAE_LIGHT_DIM, VAE_Z_DIM for clarity.