File size: 11,674 Bytes
efd4f18
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
83c1009
efd4f18
83c1009
 
efd4f18
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
83c1009
efd4f18
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
═══════════════════════════════════════════════════════════════════════════════
  H3 REALISM PEOPLE
═══════════════════════════════════════════════════════════════════════════════

───────────────────────────────────────────────────────────────────────────────
  IDENTITY
───────────────────────────────────────────────────────────────────────────────
  Base model        MiniMax H3
  Generation mode   Text-to-Video (T2VA), Image-to-Video (I2VA, first frame), Reference-to-Video (Ref2VA)
  Checkpoint        both families β€” see the author's builds
  Category          Photorealism
  Author            fal
  Source            https://huggingface.co/fal/MiniMax-H3-Realism-People-LoRA
  Provenance        HuggingFace Β· fal/MiniMax-H3-Realism-People-LoRA Β· h3-realism-people-t2v-i2v-r2v.safetensors

───────────────────────────────────────────────────────────────────────────────
  WEIGHTS
───────────────────────────────────────────────────────────────────────────────
  File              h3-realism-people-t2v-i2v-r2v.safetensors
  Version           main
  Size              125.2 MB
  Format            SafeTensor / unknown precision
  SHA256            acc529601d2da117fb81179e76c56e488a3beab1171659d305f04fa3655b787e
  Published         2026-08-12

───────────────────────────────────────────────────────────────────────────────
  TRIGGER WORDS
───────────────────────────────────────────────────────────────────────────────
  β€Ί "r34l1sm"

───────────────────────────────────────────────────────────────────────────────
  RECOMMENDED SETTINGS
───────────────────────────────────────────────────────────────────────────────
  The author did not publish specific settings.
  Start at strength 0.7 and adjust; see REFERENCE.txt for the MiniMax H3
  community baseline.

───────────────────────────────────────────────────────────────────────────────
  ADOPTION & POPULARITY
───────────────────────────────────────────────────────────────────────────────
  Downloads               53,430
  Rating                  100% positive (346/346)
  Comments                0
  Published               23 days ago
  Download rate           2,312.1/day since release

  Adoption rank           #3 of 113 in this collection  (Top tier)
  Momentum rank           #3 of 113 by download rate  (Hot Β· fresh release)

  Adoption tier is the percentile of total downloads within this collection;
  momentum is the percentile of downloads-per-day since release. Momentum is
  ranked rather than measured against a fixed rate, because this base model
  is itself new and every LoRA looks fast on an absolute scale. Both numbers
  are CivitAI's, read on the scrape date at the foot of this file.

───────────────────────────────────────────────────────────────────────────────
  CHAINING / STACKING
───────────────────────────────────────────────────────────────────────────────
  Stacks with others      not stated

  The author gave no chaining guidance. The rules below are the
  general MiniMax H3 ones, not anything specific to this LoRA.

  General MiniMax H3 rules that apply here (see REFERENCE.txt):

  β€’ Load the LoRA with LoraLoaderModelOnly between UNETLoader and your
    sampler; stacking means chaining several of those nodes in series.
  β€’ Stacks normally on top of a single acceleration/turbo LoRA; keep only
    one distill LoRA active in the chain.
  β€’ The author ships builds for both checkpoint families β€” pair each build
    with its own checkpoint and never cross-pair them.

───────────────────────────────────────────────────────────────────────────────
  PROS
───────────────────────────────────────────────────────────────────────────────
  + Widely adopted β€” 53,430 downloads. [source: CivitAI stats]
  + Positively received β€” 346 thumbs-up with no down-votes. [source:
    CivitAI stats]
  + Explicit trigger word(s) β€” r34l1sm β€” so the effect can be turned on and
    off from the prompt. [source: model metadata]
  + Author documents use across several modes (T2V, I2V, REF2V). [source:
    description]
  + Compact at 125 MB β€” cheap to stack with other LoRAs. [source: file
    size]

───────────────────────────────────────────────────────────────────────────────
  CONS & CAVEATS
───────────────────────────────────────────────────────────────────────────────
  (none recorded)

───────────────────────────────────────────────────────────────────────────────
  COMPATIBILITY WARNING
───────────────────────────────────────────────────────────────────────────────
  MiniMax H3 ships two transformers that share tensor key names: the fl2va
  checkpoint (driving T2VA, I2VA, L2VA and FL2VA) and the separate ref2va
  checkpoint (driving Ref2VA). Loading a LoRA against the wrong family can
  silently degrade output instead of raising an error, so confirm the family
  above matches the checkpoint you have loaded. Likewise, LoRAs built for
  full checkpoints often throw adaln_proj shape errors on pruned ones β€” match
  the build to your checkpoint.

───────────────────────────────────────────────────────────────────────────────
  AUTHOR'S DESCRIPTION
───────────────────────────────────────────────────────────────────────────────
  A rank-32 LoRA adapter for MiniMax H3 specialised in realistic people: skin
  that keeps its texture instead of smoothing out, coherent eyes and micro-
  expressions, film-set lighting and a subtle handheld camera quality, while
  keeping H3's native synchronized audio. Trained for 1500 steps at the high-
  resolution bucket on 176 hand-curated live-action clips, and validated with
  19 same-prompt same-seed adapter-on/off pairs. Because it only touches the
  shared attention projections (standard H3 fused-QKV layout,
  diffusion_model.blocks.N.attn.qkv_proj), it loads in ComfyUI as-is and
  works across text-, image- and reference-to-video. Successor to
  fal/MiniMax-H3-Realism-LoRA.

───────────────────────────────────────────────────────────────────────────────
  EXAMPLE MEDIA
───────────────────────────────────────────────────────────────────────────────
  1 preview(s) in ./example_images/ β€” community results for
  this LoRA. Videos keep a matching _poster.jpg still frame.

  No prompts are recorded below: CivitAI's public API flags these posts as
  having generation metadata but does not return it, so the prompts cannot be
  scraped. Open the model page to read them.

  01_video.mp4  (video, )
      still: 01_poster.jpg

───────────────────────────────────────────────────────────────────────────────
  HOW TO USE (COMFYUI)
───────────────────────────────────────────────────────────────────────────────
  1. Copy the .safetensors file into ComfyUI/models/loras/
  2. Load the matching H3 checkpoint with UNETLoader.
  3. Insert LoraLoaderModelOnly between the loader and your sampler.
  4. Set strength as listed above, then add any trigger word to the prompt.
  5. Keep CFG at 1.0 with BasicGuider unless you know you want real CFG.

───────────────────────────────────────────────────────────────────────────────
  Scraped by lorakit on 2026-09-04 23:23 UTC.
  Stats and description are the author's; pros/cons are derived from the
  evidence tagged beside each line.
───────────────────────────────────────────────────────────────────────────────