Update README.md

README.md

@@ -13,7 +13,27 @@ license: other
 license_link: LICENSE.md
 ---
 
+This is a direct GGUF conversion of the LTXV-13B-0.9.8-dev variant from [Lightricks/LTX-Video](https://huggingface.co/Lightricks/LTX-Video).
 
-# LTXV-13B-0.9.8-dev-GGUF
+The model files can be used in [ComfyUI](https://github.com/comfyanonymous/ComfyUI/) with the [ComfyUI-GGUF](https://github.com/city96/ComfyUI-GGUF) custom node. Place the required model(s) in the following folders:
 
-From Q2_K to Q8 are uploaded, we still uploading quants; once they are ready, we will update the model card.
+| Type         | Name                               | Location                          | Download 
+| ------------ | ---------------------------------- | --------------------------------- | -------------------------
+| Main Model   | LTXV-13B-0.9.8-dev                 | `ComfyUI/models/diffusion_models` | GGUF (this repo) 
+| Text Encoder | T5-V1.1-XXL-Encoder                | `ComfyUI/models/text_encoders`    | [Safetensors](https://huggingface.co/comfyanonymous/flux_text_encoders/tree/main) / [GGUF](https://huggingface.co/city96/t5-v1_1-xxl-encoder-gguf) |
+| VAE          | LTXV-13B-0.9.8-dev-VAE.safetensors | `ComfyUI/models/vae`              | [Safetensors](https://huggingface.co/QuantStack/LTXV-13B-0.9.8-dev-GGUF/resolve/main/vae/LTXV-13B-0.9.8-dev-VAE.safetensors?download=true) |
+
+The old example workflow from last version should still work, youll just need to switch the vae and the gguf.
+[**Example workflow**](https://huggingface.co/wsbagnsv1/ltxv-13b-0.9.7-distilled-GGUF/blob/main/exampleworkflow.json) - based on the [official example workflow](https://github.com/Lightricks/ComfyUI-LTXVideo/tree/master/example_workflows/)
+
+### Notes
+
+*As this is a quantized model not a finetune, all the same restrictions/original license terms still apply.*
+
+*Comfyui now supports the ggufs natively, so you just need to update comfyui to the latest version and if some issues persist update all the nodes in the workflow*
+
+*Other T5 clips will probably work as well, just use one that you like, you can get them as safetensors or ggufs. The best I tried was the t5 v1.1 xxl one*
+
+*Loras do work, but you need to follow the steps in the example workflow and dont use torchcompile with loras!*
+
+*Teacache works with ltx but atm not really good. The rel_l1_thresh only seems to work at 0.01 in my testing and even that causes some noticable quality drops, best leave it disabled.*