[Request] Workflow/Example for hy15_1080p_sr (Super Resolution) model

[Ok2dot0]

I am trying to use the new Super Resolution model from the LightX2V HuggingFace collection:
hy15_1080p_sr_cfg_distiled_fp8_e4m3_lightx2v.safetensors

I have successfully set up the wrapper for standard Text-to-Video (T2V) and Image-to-Video (I2V) tasks using the provided examples (Wan/Hunyuan). However, I cannot find any documentation or example workflows explaining how to use this specific SR (Super Resolution) model to upscale videos or images.

Goal
I want to load this sr model and use it to upscale/enhance a low-resolution input (image or video).

Questions

1. Which specific loader nodes should be used for this SR model? (Does it use the standard Lightx2vWanVideoModelLoader?)
2. In the LightX2V Inference Config node, what task_type should be selected? (Is there a hidden "sr" option, or is it the "s2v" option, or should I use "i2v" with specific settings?)
3. Could you please provide a simple JSON workflow example for 1080p upscaling using this distilled FP8 model?

Environment

• Repository: ModelTC/ComfyUI-Lightx2vWrapper
• Model: hy15_1080p_sr_cfg_distiled_fp8_e4m3_lightx2v.safetensors

Thank you for your hard work on this wrapper!

[zwukong]

Can you show us how to use hy15 in this project, i tried but failed, too many things to modify