Update readme

README.md

@@ -26,6 +26,12 @@ Here we feature some of the best:
 * Real-time generation as you type, lightning-fast: [Demo1](https://fastsdxl.ai/), [Demo2](https://huggingface.co/spaces/radames/Real-Time-Text-to-Image-SDXL-Lightning)
 * Comparison to other models: [Link](https://anotherjesse.com/posts/sdxl-lightning/)
 
+## Checkpoints
+
+* `sdxl_lightning_Nstep.safetensors`: All-in-one checkpoint, used by ComfyUI.
+* `sdxl_lightning_4step_unet.safetensors`: UNet checkpoint only, used by Diffusers.
+* `sdxl_lightning_4step_lora.safetensors`: LoRA checkpoint, for using non-SDXL base model, used by Diffusers and ComfyUI.
+
 ## Diffusers Usage
 
 Please always use the correct checkpoint for the corresponding inference steps.
@@ -110,7 +116,7 @@ pipe("A girl smiling", num_inference_steps=1, guidance_scale=0).images[0].save("
 Please always use the correct checkpoint for the corresponding inference steps.
 Please use Euler sampler with sgm_uniform scheduler.
 
-### 2-Step, 4-Step, 8-Step UNet
+### 2-Step, 4-Step, 8-Step Full
 
 1. Download the full checkpoint (`sdxl_lightning_Nstep.safetensors`) to `/ComfyUI/models/checkpoints`.
 1. Download our [ComfyUI full workflow](comfyui/sdxl_lightning_workflow_full.json).