Guide to Downloading and Setting Up Text Encoder Models for ComfyUI
If you've previously downloaded Flux (Unet), setting up the new model is easier than ever! You only need to download the model and the CLIP_G file, then place them in the clip folder to run, as they share the same set of files.
Step 1: Download the Text Encoder Files
First, ensure you have the necessary text encoder files from SD3, Flux, or other models. You’ll need the following files in your ComfyUI/models/clip/ directory:
clip_l.safetensors
clip_g.safetensors
t5xxl
For the t5xxl file, if you have more than 32GB of RAM, I recommend using t5xxl_fp16.safetensors. If not, opt for t5xxl_fp8_e4m3fn_scaled.safetensors for optimal performance.
Step 2: Download the SD3.5 Model
Place the sd3.5_large.safetensors file in your ComfyUI/models/checkpoints/ directory. Note that this file does not contain text encoder/CLIP weights, so you'll need to load them separately. Here are the links for easy downloading:
clip_g.safetensors: Download from Hugging Face
clip_l.safetensors: Download from Hugging Face
t5xxl_fp16.safetensors: Download from Hugging Face
Installation Directory
Make sure all the above files are placed in the correct ComfyUI/models/clip/ directory. For further reference, you can check here: Stable Diffusion 3.5 Large Directory.
Good luck with your installation, and enjoy it (or not)
Useful and FREE resources:
❤️Free server to make art with Flux: Shakker and Tensor Art and Sea Art
✨ More FLUX LORA? List and detailed description of each LORA I implement here: https://maitruclam.com/lora📚
🆕 First time using FLUX? Explanation and tutorial with A1111 forge offline and Comfy UI here: https://maitruclam.com/flux-ai-la-gi/🌐
🛠️ How to train your LORA with Flux? My detailed instructions are here: https://maitruclam.com/training-flux/📚
❤️ Donate me (I would be really surprised if you did that! 😄): https://maitruclam.com/donate
Find me / Contact for work on:
📱 Facebook: @maitruclam4real
💬 Discord: @maitruclam
🌐 Web: maitruclam.com
