Model Information
Description
The t2v version has been updated and also supports i2v workflow.
The t2v version has larger walking movements and faster speeds
The i2v version has a slower walking speed
The T2V version also adds some hand movements, such as:
adjusted her hair with handOr
hands on hipsMy apologies!I forgot to convert Lora to Comfy format and can only run it on Kijai's workflow. Some people are using Comfy org's gguf workflow, so this Lora cannot run properly.
I have re uploaded the converted version, please download it again, thx
I tried using Wan2.1 to generate videos of characters walking behind their backs but was unsuccessful. There were always instances of people walking backwards or making strange movements. So I trained this Lora, which will improve the success rate of generating walking back videos.
Trigger word:
walking from behindSpecial note: Do not write the prompt words too long, simply describe the character. The training video is 3 seconds (48 frames), so the duration of the generated video should not exceed 3 seconds (49 frames) as it may cause strange movements in the excess part. Perhaps increasing the number of sampling steps can lead to improvement
The Walking Back - Wan I2V 14B
i2v_v2.0
Model Details
- Type
- Model Addon
- Subtype
- LoRA
- Created
- Updated
- April 12, 2026