Update README.md

README.md

@@ -29,12 +29,25 @@ inference: true
 
 I hope that Blacklight Makeup's radiant glow inspires you to experiment, to challenge norms, and to create beauty that transcends the ordinary!
 
-### INSTRUCTIONS:
+### What's new in Version 2?
+
+I've retrained it with *improved captions and parameters*, which brings:
+
+- simpler trigger words: `blacklight makeup`
+- better output quality
+- reduced file size
+- improved compatibility with other LoRAs
+
+### What's next?
+
+Enhancing the dataset while also experimenting with new training techniques.
+
+### How to use:
 
 **Example prompt:** `Portrait photo of a kitty cat with blacklight makeup, fierce eyes, light bokeh, sharp focus, soft lighting, detailed, intricate, elegant, vibrant colors, masterpiece, cinematic, 8k`
 
-- trigger words: `with blacklight makeup`
-- `<lora:blacklight_makeup_v1:1>`
+- trigger words: `blacklight makeup`
+- `<lora:blacklight_makeup_v2:1>`
 - **2:3** — 832x1248
 - **16:9** — 1360x768
 - **1:1** — 1024x1024
@@ -58,12 +71,15 @@ pipe.scheduler = EulerDiscreteScheduler.from_config(
 
 pipe.to("cuda")
 
-pipe.load_lora_weights("chillpixel/blacklight-makeup-sdxl-lora")
+pipe.load_lora_weights(
+    "chillpixel/blacklight-makeup-sdxl-lora",
+    weight_name="blacklight_makeup_v2.safetensors",
+)
 
 image = pipe(
-    prompt="portrait photo of a kitty cat with blacklight makeup, fierce eyes, light bokeh, sharp focus, soft lighting, detailed, intricate, elegant, vibrant colors, masterpiece, cinematic, 8k",
-    num_inference_steps=40,
-    guidance_scale=8,
+    prompt="Portrait of woman with blacklight makeup, fantasy, highly detailed, digital painting, artstation, concept art, sharp focus, illustration, art by Tony Sart and artgerm and randy vargas",
+    num_inference_steps=35,
+    guidance_scale=6,
     width=832,
     height=1248,
 ).images[0]