ComfyUI Node
WanAnimatePlus TextEncodeCached
Encodes text prompts into text embeddings. This node loads and completely unloads the T5 after done, leaving no VRAM or RAM imprint. If prompts have been cached before T5 is not loaded at all. negative output is meant to be used with NAG, it contains only negative prompt embeddings. Additionally you can provide a Qwen LLM model to extend the positive prompt with either one of the original Wan templates or a custom system prompt.
WanAnimatePlus TextEncodeCached
- extender_args
- text_embeds
- negative_text_embeds
- positive_prompt
◄model_name▾►
◄precisionbf16►
◄positive_prompt►
◄negative_prompt►
◄quantizationdisabled►
◄use_disk_cachetrue►
◄devicegpu►
CategoryWanAnimatePlus
Inputs (8)
| Name | Type | Default | Description |
|---|---|---|---|
| model_name | COMBO | These models are loaded from 'ComfyUI/models/text_encoders' | |
| precision | COMBO | bf16 | 2 options: fp32, bf16 |
| positive_prompt | STRING | — | |
| negative_prompt | STRING | — | |
| quantization | COMBO | disabled | optional quantization method |
| use_disk_cache | BOOLEAN | true | Cache the text embeddings to disk for faster re-use, under the custom_nodes/ComfyUI-WanVideoWrapper/text_embed_cache directory |
| device | COMBO | gpu | Device to run the text encoding on. |
| extender_argsopt | WANVIDEOPROMPTEXTENDER_ARGS | Use this node to extend the prompt with additional text. |
Outputs (3)
| Name | Type | Description |
|---|---|---|
| text_embeds | WANVIDEOTEXTEMBEDS | The text embeddings for both prompts |
| negative_text_embeds | WANVIDEOTEXTEMBEDS | The text embeddings for the negative prompt only (for NAG) |
| positive_prompt | STRING | Positive prompt to display prompt extender results |