Update README.md

README.md

@@ -1,10 +1,22 @@
 ---
 license: openrail++
 ---
-<b>RealVisXL V5.0 Status (Updated: Aug 31, 2024):<br>
-- Training Images: +3340 (V4.0: 3340)<br>
-- Training Steps: +668k (V4.0: 672k)<br>
-- Approximate percentage of completion: ~100%<br>
-- Training Time: ~236.6h (V1.0 - V3.0: ~202h)<br></b>
+<b>This model is available on ---</b><br>
+<b>You can support me directly on Boosty - https://boosty.to/sg_161222</b><br>
 
-<a href="https://boosty.to/sg_161222/posts/58d263ff-dded-4868-99bd-2f85146083c0">Latest Post About RealVisXL V5.0</a>
+The model is aimed at photorealism. Can produce sfw and nsfw images of decent quality.<br>
+CivitAI Page: ---<br>
+
+<b>Recommended Negative Prompt:</b><br>
+bad hands, bad anatomy, ugly, (face asymmetry, eyes asymmetry, deformed eyes, open mouth)<br>
+<b>or another negative prompt</b><br>
+
+<b>Recommended Generation Parameters:</b><br>
+Sampling Method: DPM++ SDE Karras (30+ Sampling Steps) or DPM++ 2M Karras (50+ Sampling Steps)<br>
+
+<b>Recommended Hires Fix Parameters:</b><br>
+Hires Sampling Method: DPM++ 2M Karras<br>
+Hires steps: 25+<br>
+Upscaler: 4x-NMKD-Superscale-SP_178000_G / 4x-UltraSharp upscaler / or another<br>
+Denoising strength: 0.1 - 0.3<br>
+Upscale by: 1.1-2.0<br>