This work flow integrates AnimagineXL 3.1 output into CogvideoX 1.0 Img2Vid flow. You can use the "enhance prompt" feature in create mode for the cogvideo flow prompt. Most of the time, it works okay, but will give you out of memory error if you're unlucky.
Animagine XL 3.1 + LoRa + CogVideoX 1.0 I2V #Christmas Walkthrough
Workflow Preview
Showcases (Image/Video)

Showcases (Image/Video)

Nodes Detail
22 NodesPrimitive Node Types
7- PreviewImage1
- VAEDecode1
- KSampler1
- CLIPTextEncode2
- EmptyLatentImage1
- CLIPLoader1
Custom Node Types
15- CogVideoTextEncode2
- CogVideoImageEncode1
- CogVideoDecode1
- TensorArt_LoraLoader3
- VHS_VideoCombine1
- DownloadAndLoadCogVideoModel1
- TensorArt_CheckpointLoader1
- TensorArt_PromptText2
- TensorArt_Seed1
- ImageResizeKJ1
- CogVideoSampler1