This is the Lora taken directly from Kijai's Huggingface - https://huggingface.co/Kijai/WanVideo_comfy/blob/main/Lightx2v/lightx2v_T2V_14B_cfg_step_distill_v2_lora_rank32_bf16.safetensors
I'm just adding here so I can use it in my metadata
Kijai has a lot more loras that can be useful to you, please check them in his official repo
https://huggingface.co/Kijai/WanVideo_comfy/tree/main
Been using this one with Wan2.2 with the following strength to generate images, but it should work for videos:
High - 3
Low - 1.5
Having good results with 6 steps ( 3 in each )
Description
FAQ
Details
Downloads
2,107
Platform
CivitAI
Platform Status
Available
Created
9/6/2025
Updated
4/27/2026
Deleted
-
Files
lightx2v_14B_T2V_cfg_step_distill_lora_adaptive_rank_quantile_0.15_bf16.safetensors
Mirrors
Huggingface (3 mirrors)
CivitAI (1 mirrors)
Other Platforms (TensorArt, SeaArt, etc.) (1 mirrors)
lightx2v_t2v_14b_cfg_step_distill_v2_lora_rank32_bf16.safetensors
Mirrors
Huggingface (4 mirrors)
CivitAI (2 mirrors)
Other Platforms (TensorArt, SeaArt, etc.) (1 mirrors)
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.