feat: rename ResTool32xLTX to ResTool64xLTX for IC-LoRA compatibility

- Renamed `ResTool32xLTX` to `ResTool64xLTX` and updated DIVISOR to 64 to prevent odd-dimension `ValueError`s when using latent downscale factors of 2.0 (IC-LoRA).
- Updated visual node name in `__init__.py` to accurately reflect the 64x divisor behavior to avoid technical debt.
- Updated documentation in `README.md` and `README_EN.md` to reflect the rename and explain the downscale compatibility logic.
This commit is contained in:
Meisoftcoltd
2026-04-03 13:06:59 +00:00
parent 1ffccc5599
commit e443ec7112
4 changed files with 5 additions and 5 deletions
+1 -1
View File
@@ -13,7 +13,7 @@ Una suite de grado profesional de nodos personalizados para ComfyUI. Diseñada p
* **Cortes Inteligentes (Smart Chunking):** Analiza el vídeo base y realiza cortes matemáticos evitando separar frames donde los rostros son más nítidos, manteniendo la coherencia de identidad.
* **Extracción de Audio Nativa:** Extrae la pista de audio original directamente desde el nodo inicial y la inyecta de vuelta en el ensamblaje final.
* **Dual Input Engine:** El nodo explorador acepta tanto vídeos subidos manualmente como rutas inyectadas por cable desde descargadores (ej. YTDLP) comportándose exactamente como VideoHelperSuite.
* **Soporte Nativo LTX 2.3:** Se han añadido calculadores matemáticos específicos (`AutoLoopCalculatorLTX` y `ResTool32xLTX`) para garantizar que los lotes de vídeo cumplan con las estrictas reglas de descompresión de arquitectura DiT (resoluciones divisibles por 32 y particiones de fotogramas de `8n + 1`).
* **Soporte Nativo LTX 2.3:** Se han añadido calculadores matemáticos específicos (`AutoLoopCalculatorLTX` y `ResTool64xLTX`) para garantizar que los lotes de vídeo cumplan con las estrictas reglas de descompresión de arquitectura DiT y el factor de escala latente (resoluciones divisibles por 64 y particiones de fotogramas de `8n + 1`).
---
+1 -1
View File
@@ -13,7 +13,7 @@ A professional-grade suite of custom nodes for ComfyUI. Designed to bypass VRAM
* **Smart Chunking:** Analyzes the base video and performs mathematical cuts avoiding the separation of frames where faces are sharpest, thus maintaining identity coherence.
* **Native Audio Extraction:** Extracts the original audio track straight from the entry node and seamlessly injects it back into the final stitch.
* **Dual Input Engine:** The explorer node accepts both manually uploaded videos and dynamically injected string paths from downloaders (e.g., YTDLP), behaving exactly like VideoHelperSuite.
* **Native LTX 2.3 Support:** Specific mathematical calculators (`AutoLoopCalculatorLTX` and `ResTool32xLTX`) have been added to ensure video batches comply with the strict DiT architecture decompression rules (resolutions strictly divisible by 32 and frame partitions of `8n + 1`).
* **Native LTX 2.3 Support:** Specific mathematical calculators (`AutoLoopCalculatorLTX` and `ResTool64xLTX`) have been added to ensure video batches comply with the strict DiT architecture decompression rules and latent downscale factors (resolutions strictly divisible by 64 and frame partitions of `8n + 1`).
---
+1 -1
View File
@@ -26,8 +26,8 @@ NODE_DISPLAY_NAME_MAPPINGS.update({
"ResTool8x": "📐 ResTool 8x (SD1.5)",
"ResTool16x": "📏 ResTool 16x (SDXL)",
"ResTool32x": "🎞️ ResTool 32x (WanVideo)",
"ResTool32xLTX": "🌌 ResTool 32x (LTX 2.3)",
"ResTool64x": "🎬 ResTool 64x (Hunyuan)",
"ResTool64xLTX": "🌌 ResTool 64x (LTX 2.3)",
})
__all__ = ['NODE_CLASS_MAPPINGS', 'NODE_DISPLAY_NAME_MAPPINGS']
+2 -2
View File
@@ -85,8 +85,8 @@ class ResTool64x(BaseResolutionTool):
MIN_PIXELS = 921600 # Hunyuan (Equivale a 1280x720)
@register_node
class ResTool32xLTX(BaseResolutionTool):
DIVISOR = 32
class ResTool64xLTX(BaseResolutionTool):
DIVISOR = 64
MIN_PIXELS = 393216 # LTX 2.3 (Equivale a 768x512)