CivArchive
Model Type
Base Model
Platform
Add Mirror
Natural Face & Speech H3/LTX 2.5 - v1.0-LTX2.5
NSFW
v1.0-LTX2.5
v2.0-MH3
v1.0-MH3
Natural Face & Speech for H3/LTX 2.5
Full face muscle dynamics with clear English speech
Trained on T2V but supports also I2V
720x1280 resolution with 15-30 steps
Weight 0.6-0.8 is recommended
⚠️
Avoid prompt bleeding using shorter sentences or words
Description
First initial release for LTX 2.5
FAQ
What is Natural Face & Speech H3/LTX 2.5?
How do I use Natural Face & Speech H3/LTX 2.5?
Why might this LoRA not be producing the expected results?
Can I use this LoRA commercially?
What files are available and where can I download them?
LORA
LTXV 2.5
by
AdaptiveVision
Download (Beta)
View on CivitAI
minimax
i2v
t2v
lora
h3
concept
Details
Downloads
75
Platform
CivitAI
Platform Status
Available
Created
11/4/2026
Updated
11/5/2026
Deleted
-
Files
nafasp_ltx_25_500.safetensors
Size:
408.32 MB
SHA256:
0ff35e80286f845cc5ab9271a8310c3fe0dd8f1e9cecef7a728eb6a9629a4c20
Mirrors
CivitAI (1 mirrors)
nafasp_ltx_25_500.safetensors