LTXV 13B Distilled + Upscale:

This ComfyUI workflow uses the LTXV 13B 0.9.7 Distilled model for image-to-video (i2v) generation, with support for both base and tiled samplers, and includes an integrated latent upscaling pipeline for enhanced quality.

Key Features:

• LTXV Base Sampler with STG Guider Advanced, supporting precise control via sigma/cfg/stg tuning.

• Optional Image Conditioning, allowing i2v prompts with visual input anchors.

• Latent Upscaler using the official spatial upscaler model (0.9.7), with optional AdaIN blending from reference latents.

• Tiled Sampler Branch for higher resolutions and extended generation scenarios.

• Integrated VAE decode (standard and tiled) and VHS Video Combine for direct video output.

• Optional Film Grain & Resize stage for cinematic finishing.

This setup is suitable for users aiming to generate higher-quality outputs with the distilled LTXV model, using either low-res base output or tiled+upscaled paths, and is fully compatible with LoRA and conditioning workflows.