ComfyUI Node
Wan Video Image To Video Encode_v2 (QQ)
A ComfyUI node in WanVideoWrapper_QQ/utils with 22 inputs and 1 output.
Wan Video Image To Video Encode_v2 (QQ)
- vae
- clip_embeds
- start_image
- mid_image
- end_image
- control_embeds
- temporal_mask
- extra_latents
- add_cond_latents
- image_embeds
◄width832►
◄height480►
◄num_frames81►
◄noise_aug_strength0.000►
◄start_latent_strength1.000►
◄mid_latent_strength1.000►
◄end_latent_strength1.000►
◄end_final_strength0.800►
◄mid_position0.50►
◄end_position1.00►
◄force_offloadtrue►
◄fun_or_fl2v_modeltrue►
◄tiled_vaefalse►
CategoryWanVideoWrapper_QQ/utils
Inputs (22)
| Name | Type | Default | Description |
|---|---|---|---|
| width | INT | 83264–8096 | Width of the image to encode |
| height | INT | 48064–8096 | Height of the image to encode |
| num_frames | INT | 811–10000 | Number of frames to encode |
| noise_aug_strength | FLOAT | 0.0000–10 | Strength of noise augmentation, helpful for I2V where some noise can add motion and give sharper results |
| start_latent_strength | FLOAT | 1.0000–10 | Additional latent multiplier, helpful for I2V where lower values allow for more motion |
| mid_latent_strength | FLOAT | 1.0000–10 | Additional latent multiplier for mid frame, helpful for I2V where lower values allow for more motion |
| end_latent_strength | FLOAT | 1.0000–10 | Additional latent multiplier, helpful for I2V where lower values allow for more motion |
| end_final_strength | FLOAT | 0.8000–10 | Strength for end_image copy placed at final frame for temporal consistency |
| mid_position | FLOAT | 0.500–1 | Position of mid_image as fraction of total frames (0.0 = start, 1.0 = end) |
| end_position | FLOAT | 1.000–1 | Position of end_image relative to remaining timeline after mid_image |
| force_offload | BOOLEAN | true | — |
| vaeopt | WANVAE | — | |
| clip_embedsopt | WANVIDIMAGE_CLIPEMBEDS | Clip vision encoded image | |
| start_imageopt | IMAGE | Image to encode | |
| mid_imageopt | IMAGE | middle frame | |
| end_imageopt | IMAGE | end frame | |
| control_embedsopt | WANVIDIMAGE_EMBEDS | Control signal for the Fun -model | |
| fun_or_fl2v_modelopt | BOOLEAN | true | Enable when using official FLF2V or Fun model |
| temporal_maskopt | MASK | mask | |
| extra_latentsopt | LATENT | Extra latents to add to the input front, used for Skyreels A2 reference images | |
| tiled_vaeopt | BOOLEAN | false | Use tiled VAE encoding for reduced memory use |
| add_cond_latentsopt | ADD_COND_LATENTS | Additional cond latents WIP |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| image_embeds | WANVIDIMAGE_EMBEDS | — |