Sentinel7 commited on
Commit
a4b4434
·
verified ·
1 Parent(s): 0e11f0c

Upload 2856467/3416478/README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. 2856467/3416478/README.md +88 -0
2856467/3416478/README.md ADDED
@@ -0,0 +1,88 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: other
3
+ tags:
4
+ - breasts
5
+ - concept
6
+ - nipples
7
+ - nsfw
8
+ - penis
9
+ - sex
10
+ - vagina
11
+ - woman
12
+ ---
13
+
14
+ # [MMH3] Mystic XXX - v5.0 (FL2VA & REF2VA) - 3416478
15
+
16
+ **Model Type**: LORA
17
+
18
+ **Base Model**: MiniMax H3
19
+
20
+ **Trigger Words**: None
21
+
22
+ **Tags**: breasts, concept, nipples, nsfw, penis, sex, vagina, woman
23
+
24
+ ## Gallery
25
+
26
+ <table>
27
+ <tr>
28
+ <td><video src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/0bec3459-ac47-4f92-8cb3-4454b559955f/original=true/146187660.mp4" width="200" controls muted autoplay loop></video></td>
29
+ <td><video src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/10dbfdff-10f9-4142-b837-0cb24e8a33ab/original=true/146188408.mp4" width="200" controls muted autoplay loop></video></td>
30
+ <td><video src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/ce5169e1-9f8a-4d6b-987c-52e97def3ba5/original=true/146188476.mp4" width="200" controls muted autoplay loop></video></td>
31
+ </tr>
32
+ <tr>
33
+ <td><video src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/e28905f6-57ce-436f-b468-e593de73d314/original=true/146188551.mp4" width="200" controls muted autoplay loop></video></td>
34
+ <td><video src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/596f0cf3-2f07-432f-913d-f413358ef231/original=true/146188569.mp4" width="200" controls muted autoplay loop></video></td>
35
+ <td><video src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/23346a61-9f9e-4301-beda-4c77f7eae944/original=true/146188610.mp4" width="200" controls muted autoplay loop></video></td>
36
+ </tr>
37
+ <tr>
38
+ <td><video src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/49ab707f-86a0-417c-a52a-657afd341532/original=true/146188646.mp4" width="200" controls muted autoplay loop></video></td>
39
+ <td><video src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/f1e54393-b5fb-4300-9391-c70d9ebafe46/original=true/146188767.mp4" width="200" controls muted autoplay loop></video></td>
40
+ <td><video src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/419a493b-54e2-4672-8d3b-96d5fe7a240d/original=true/146260740.mp4" width="200" controls muted autoplay loop></video></td>
41
+ </tr>
42
+ <tr>
43
+ <td><video src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/84ca34ee-30e6-4e6b-876f-bb249c94723c/original=true/146260776.mp4" width="200" controls muted autoplay loop></video></td>
44
+ </tr>
45
+ </table>
46
+
47
+ ## Description
48
+
49
+ V5 - REF2VA
50
+ ===========
51
+
52
+ This version was refined from the FL2VA using only MLP from 10-49 to reduce interference with other loras and references, it has the same audio as FL2VA and was retrained to gain reference capabilities that FL2VA don't have, for this a dataset with references and goals is used. It was made with Musubi Trainer.
53
+
54
+ V5 - FL2VA
55
+ ==========
56
+
57
+ This version was trained with audio using the full layers spectrum, including both MLP and attention layers, with Musubi Trainer.
58
+
59
+ Compared to V4, it delivers more natural skin rendering and lighting, improved movement, more realistic results, and better audio quality. The dataset was increased by approximately 50%, and the training resolution was raised to capture more detail. Anatomy is also slightly improved, with better prompt adherence in some cases.
60
+
61
+ I tested it against V4 across more than 100 videos, and this version produced better results in approximately 70% of them. The validation results also improved, with a particularly significant improvement in T2VA performance.
62
+
63
+ V5 is based on FL2VA and uses all layers, resulting in a larger LoRA with a greater potential for interference with other LoRAs, particularly those targeting overlapping layers. If you encounter compatibility issues, let me know. A version trained without attention and identity layers may be a viable alternative, although removing these layers would likely reduce the overall quality.
64
+
65
+ ---
66
+
67
+ New ref2va file to download (experimental, no audio trained)
68
+ ============================================================
69
+
70
+ ![](https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/1ef67ea7-2238-4d0f-8255-4059b7e20d27/original=true/1ef67ea7-2238-4d0f-8255-4059b7e20d27.jpeg)
71
+
72
+ **Mystic XXX – Unlock Real Anatomy in T2V**
73
+
74
+ Finally, text-to-video with proper anatomy. Works great with **T2V, I2V and FFLF**.
75
+
76
+ All example videos come with the full workflow embedded. Recommended strength: **0.2 – 1** (V4 I run at 1). Turbo LoRAs can change the look, adjust strength as needed.
77
+
78
+ Download, try it, and have fun!
79
+
80
+ ---
81
+
82
+ Author: [alcaitiff](https://civitai.com/user/alcaitiff)
83
+
84
+ Model: [CivitAI Model Page](https://civitai.com/models/2856467?modelVersionId=3416478)
85
+
86
+ Archive: [CivArchive Page](https://civarchive.com/models/2856467?modelVersionId=3416478)
87
+
88
+ <!-- Version: 20261020_upload -->