# ComfyUI節點
## CLIP
> Contrastive Language-Image Pre-Training
- CLIP model: 連接文本和圖像的模型
- 在load_clip, load_checkpoint的時候被初始化
- loaders -> Load Checkpoint/CheckpointLoaderSimple
- loaders -> unCLIPCheckpointLoader
- advanced -> loaders -> Load CLIP/CLIPLoader
- advanced -> loaders -> DualCLIPLoader
- advanced -> loaders -> Load Checkpoint With Config (DEPRECATED)/CheckpointLoader
### CLIPTextEncode
[文件](https://blenderneko.github.io/ComfyUI-docs/Core%20Nodes/Conditioning/CLIPTextEncode/)
- conditioning -> CLIP Text Encode (Prompt)
- 功能: 轉換成conditioning
### CLIPSetLastLayer
[文件](https://blenderneko.github.io/ComfyUI-docs/Core%20Nodes/Conditioning/CLIPSetLastLayer/)
- conditioning -> CLIP Set Last Layer
- 功能: 設定CLIP model的output layer
### LoraLoader
[文件](https://blenderneko.github.io/ComfyUI-docs/Core%20Nodes/Loaders/LoadLoRA/)
- loaders -> Load LoRA
- 功能: 以lora model來調整diffusion model和CLIP model
### GLIGENTextBoxApply
[文件](https://blenderneko.github.io/ComfyUI-docs/Core%20Nodes/Conditioning/GLIGENTextboxApply/)
- conditioning -> gligen -> GLIGENTextBoxApply
- 目前僅支援SD1.5，不支援SDXL
- 功能: 用於指定圖像生成的大小、位置和空間關係