Rin247 commited on
Commit
8e26653
·
verified ·
1 Parent(s): 502cf36

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +43 -1
README.md CHANGED
@@ -12,7 +12,7 @@ library_name: diffusers
12
 
13
  # Anima-Turbo-Diffusers
14
 
15
- Quantised media packages of `Anima-Turbo-Diffusers`.
16
 
17
  ## MECHA Forge
18
 
@@ -21,6 +21,10 @@ MECHA — Model Engineering & Contextual Hearth Assembly
21
  These media packages have been forged and optimized in the MECHA Forge.
22
  Each one holds the Denoiser, Text Encoder, and VAE components packaged together for easier deployment.
23
 
 
 
 
 
24
  ## Available quantizations
25
 
26
  | Quant | Denoiser | Text encoder | VAE |
@@ -31,3 +35,41 @@ Each one holds the Denoiser, Text Encoder, and VAE components packaged together
31
  | INT8 | INT8 | INT8 | INT8 |
32
 
33
  The VAE is shared between formats: FP4 reuses the FP8 file, INT4 reuses the INT8 file.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
12
 
13
  # Anima-Turbo-Diffusers
14
 
15
+ Quantised media packages of [circlestone-labs/Anima](https://huggingface.co/circlestone-labs/Anima) (Anima 1.0 Turbo, v1.1 weights).
16
 
17
  ## MECHA Forge
18
 
 
21
  These media packages have been forged and optimized in the MECHA Forge.
22
  Each one holds the Denoiser, Text Encoder, and VAE components packaged together for easier deployment.
23
 
24
+ ## Description
25
+
26
+ Anima 1.0 Turbo (v1.1 weights): distilled 2B-parameter anime/illustration text-to-image model (Cosmos-based). Fast generation with increased stability and a strong default style, at some cost to diversity. Uses a Qwen3-0.6B text encoder and the Qwen-Image VAE.
27
+
28
  ## Available quantizations
29
 
30
  | Quant | Denoiser | Text encoder | VAE |
 
35
  | INT8 | INT8 | INT8 | INT8 |
36
 
37
  The VAE is shared between formats: FP4 reuses the FP8 file, INT4 reuses the INT8 file.
38
+
39
+ ## Recommended settings
40
+
41
+ From the source model card (adjust to taste):
42
+
43
+ | Parameter | Value |
44
+ |---|---|
45
+ | Steps | 8-12 |
46
+ | CFG / Guidance scale | 1.0 |
47
+ | Resolution | 512-1536 (total pixels) |
48
+ | Notes | Distilled; few-step sampling; CFG 1 |
49
+
50
+ ## Example prompts
51
+
52
+ Caught in the moment (candid, slice of life):
53
+
54
+ ```
55
+ masterpiece, best quality, score_7, safe, 1girl, solo, long hair, blue
56
+ eyes, white blouse, pleated skirt, looking at viewer, gentle smile,
57
+ smiling, city lights at night, soft bokeh, anime illustration.
58
+ ```
59
+
60
+ Fashion photography (magazine cover style):
61
+
62
+ ```
63
+ masterpiece, best quality, score_7, safe, 1boy, solo, short dark hair,
64
+ sharp jawline, tailored navy suit, open-collar white shirt, studio rim
65
+ lighting, bold serif typography, high contrast, cinematic, 8K, sharp
66
+ focus, GQ editorial.
67
+ ```
68
+
69
+ Scenery:
70
+
71
+ ```
72
+ masterpiece, best quality, score_7, safe, scenery, misty mountain ridge
73
+ at dawn, pine forests below, low clouds drifting through the valley,
74
+ first light on snow-capped peaks, anime illustration, wide angle, 8K.
75
+ ```