Sentinel7 commited on
Commit
175efa6
·
verified ·
1 Parent(s): cce375f

Upload 2334739/2626300/README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. 2334739/2626300/README.md +55 -3
2334739/2626300/README.md CHANGED
@@ -1,10 +1,62 @@
1
  ---
2
  license: other
3
  tags:
4
- - civitai
 
 
 
 
5
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
6
  Author: [Phr00t_](https://civitai.com/user/Phr00t_)
7
 
8
- Model: [https://civitai.com/models/2334739?modelVersionId=2626300](https://civitai.com/models/2334739?modelVersionId=2626300)
 
 
9
 
10
- Mirror: [https://civarchive.com/models/2334739?modelVersionId=2626300](https://civarchive.com/models/2334739?modelVersionId=2626300)
 
1
  ---
2
  license: other
3
  tags:
4
+ - concept
5
+ - ltx2
6
+ - nsfw
7
+ - porn
8
+ - pov
9
  ---
10
+
11
+ # POV NSFW LTX2 (Updated v3.0)-v1.0 - 2626300
12
+
13
+ **Model Type**: LORA
14
+
15
+ **Base Model**: LTXV
16
+
17
+ **Trigger Words**: None
18
+
19
+ **Tags**: concept, ltx2, nsfw, porn, pov
20
+
21
+ ## Gallery
22
+
23
+ <table>
24
+ <tr>
25
+ <td><video src="https://c.genur.art/cddb94d9-b9ae-434d-8282-8487b0524e14" width="200" controls muted autoplay loop></video></td>
26
+ </tr>
27
+ </table>
28
+
29
+ ## Description
30
+
31
+ **v3.0:** 24k steps trained on a cleaner set of 132 POV videos.
32
+
33
+ Trained concepts: boobjobs, cumshots, handjobs, vaginal sex (front/back), licking and blowjobs.
34
+
35
+ Significantly improved dialog and speech issues.
36
+
37
+ I'm working on NSFW merges on my Huggingface, which will be better than using this LORA alone.
38
+
39
+ **v2.0 doesn't exist, it was a failed run.**
40
+
41
+ **v1.0:** 16k steps trained on 126 random POV pornographic video/audio data. Used ai-trainer. Rank 64 LORA.
42
+
43
+ Cumshots, handjobs, vaginal sex, boobjobs, blowjobs and some licking included in the dataset. Overall works pretty good, but not great.
44
+
45
+ Only use with I2V, which the model was trained with. First to last frame for absolute best results. T2V probably sucks.
46
+
47
+ I'm working on an actual merge with other good NSFW LORAs. See my workflows and merges here: <https://huggingface.co/Phr00t/LTX2-Rapid-Merges>
48
+
49
+ My example was using my "DoEverything" workflow with this LORA at 1.0 strength on top of my "v3 SFW" LTX2 merge.
50
+
51
+ **TRAINING TIPS:** Loss graphs are kinda garbage and I didn't really see a drop in error until after ~8k steps, and loss only went from like 1.05 to 0.8, yet results are pretty good. Short videos that focus on a specific action seem to work pretty good with clear prompting of both visuals, motion and audio. Set "caption dropout" to 0 (ai-toolkit defaults to 0.05) and use the "experimental" differential guidance set to 2.5 (at least those are the settings I used). I'd focus far more on video data than image data.
52
+ **Want to help me pay for runpod time?** Paypal donations accepted at phr00t at gmail dot com!
53
+
54
+ ---
55
+
56
  Author: [Phr00t_](https://civitai.com/user/Phr00t_)
57
 
58
+ Model: [CivitAI Model Page](https://civitai.com/models/2334739?modelVersionId=2626300)
59
+
60
+ Archive: [CivArchive Page](https://civarchive.com/models/2334739?modelVersionId=2626300)
61
 
62
+ <!-- Version: 20260424_all_update -->