SOLRICKS commited on
Commit
b91493b
·
verified ·
1 Parent(s): 69283a3

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +10 -16
README.md CHANGED
@@ -7,14 +7,15 @@ base_model:
7
  tags:
8
  - lora
9
  - character
10
- - character-consistency
 
11
  - mila
12
  - ltx-2.3
13
  - z-image-turbo
14
- - krea-2
15
  - krea-2-turbo
16
  - text-to-image
17
  - text-to-video
 
18
  - realistic
19
  - cinematic
20
  - fashion
@@ -29,27 +30,19 @@ widget:
29
  output:
30
  url: models/z-image-turbo/mila-amalfi-coast.png
31
 
32
- - text: "Mila — Positano lifestyle video"
33
  output:
34
  url: models/ltx-2.3/mila-positano-balcony.mp4
35
 
36
- - text: "Mila — fitness lifestyle video"
37
- output:
38
- url: models/ltx-2.3/mila-fitness-video.mp4
39
-
40
- - text: "Mila — fashion street video"
41
- output:
42
- url: models/ltx-2.3/mila-fashion-street-video.mp4
43
-
44
  - text: "Mila — cafe lifestyle video"
45
  output:
46
  url: models/ltx-2.3/mila-cafe-lifestyle.mp4
47
 
48
- - text: "Mila — grocery shopping lifestyle video"
49
  output:
50
  url: models/ltx-2.3/mila-grocery-shopping.mp4
51
 
52
- - text: "Mila — luxury car lifestyle video"
53
  output:
54
  url: models/ltx-2.3/mila-luxury-car.mp4
55
 
@@ -85,7 +78,7 @@ Each LoRA is trained specifically for its corresponding base model and should on
85
 
86
  **Base Model:** `Lightricks/LTX-2.3`
87
  **Trigger:** `mila`
88
- **Recommended LoRA Strength:** `1.0`
89
  **Training Steps:** `2000`
90
 
91
  The LTX-2.3 version is designed for realistic character-focused video generation while maintaining Mila's facial identity across motion, expressions, environments and camera compositions.
@@ -95,7 +88,8 @@ It works particularly well for:
95
  - Lifestyle videos
96
  - Fashion content
97
  - Travel scenes
98
- - Social media / Reels-style shots
 
99
  - Fitness scenes
100
  - Luxury environments
101
  - Natural facial expressions
@@ -114,7 +108,7 @@ realistic cinematic lifestyle video, natural movement
114
 
115
  The repository includes several LTX-2.3 generations demonstrating Mila across different environments and motion scenarios:
116
 
117
- **Positano / Coastal Lifestyle**
118
 
119
  ` samples/ltx-2.3/mila-positano-balcony.mp4 `
120
 
 
7
  tags:
8
  - lora
9
  - character
10
+ - character
11
+ - consistency
12
  - mila
13
  - ltx-2.3
14
  - z-image-turbo
 
15
  - krea-2-turbo
16
  - text-to-image
17
  - text-to-video
18
+ - image-to-video
19
  - realistic
20
  - cinematic
21
  - fashion
 
30
  output:
31
  url: models/z-image-turbo/mila-amalfi-coast.png
32
 
33
+ - text: "Mila — Positano balcony video"
34
  output:
35
  url: models/ltx-2.3/mila-positano-balcony.mp4
36
 
 
 
 
 
 
 
 
 
37
  - text: "Mila — cafe lifestyle video"
38
  output:
39
  url: models/ltx-2.3/mila-cafe-lifestyle.mp4
40
 
41
+ - text: "Mila — grocery shopping video"
42
  output:
43
  url: models/ltx-2.3/mila-grocery-shopping.mp4
44
 
45
+ - text: "Mila — luxury car video"
46
  output:
47
  url: models/ltx-2.3/mila-luxury-car.mp4
48
 
 
78
 
79
  **Base Model:** `Lightricks/LTX-2.3`
80
  **Trigger:** `mila`
81
+ **Recommended LoRA Strength:** `0.8`
82
  **Training Steps:** `2000`
83
 
84
  The LTX-2.3 version is designed for realistic character-focused video generation while maintaining Mila's facial identity across motion, expressions, environments and camera compositions.
 
88
  - Lifestyle videos
89
  - Fashion content
90
  - Travel scenes
91
+ - Social media
92
+ - Reels-style shots
93
  - Fitness scenes
94
  - Luxury environments
95
  - Natural facial expressions
 
108
 
109
  The repository includes several LTX-2.3 generations demonstrating Mila across different environments and motion scenarios:
110
 
111
+ **Positano / Coastal**
112
 
113
  ` samples/ltx-2.3/mila-positano-balcony.mp4 `
114