28 lines
1.3 KiB
Markdown
28 lines
1.3 KiB
Markdown
## TT SD3.5 Text Encoder (Context) *TT_Sd35TextEncoderContextNode*
|
|
|
|
*This description was generated with AI assistance. If you spot any errors, please create an issue on GitHub.*
|
|
|
|
Encodes text prompts for SD3.5 using all three text encoders (CLIP-L, CLIP-G, T5-XXL) from context and returns a configured Guider object.
|
|
|
|
### Inputs
|
|
|
|
| Parameter | Type | Required | Description |
|
|
|-----------|------------|----------|-------------------------------------------------------------------------|
|
|
| `context` | TT_CONTEXT | Yes | Context object containing model, CLIP, prompts, and resolution settings |
|
|
|
|
### Outputs
|
|
|
|
| Parameter | Type | Description |
|
|
|-----------|------------|---------------------------------------------------------|
|
|
| `CONTEXT` | TT_CONTEXT | Updated context with guider stored |
|
|
| `GUIDER` | GUIDER | Configured guider with encoded conditioning for sampler |
|
|
|
|
### Usage
|
|
|
|
Connect `CONTEXT` from a SD3.5 workflow settings context node. Connect `GUIDER` to a sampler node. Prompts and resolution are read from context — use
|
|
`TT SD3.5 GGUF Workflow Settings (Advanced)` to set per-encoder prompts.
|
|
|
|
---
|
|
|
|
[View on GitHub](https://github.com/tenser-tensor/ComfyUI-TenserTensor)
|