Download
1 variant available
7,3100 1 2 3 4 5 6 7 8 9,0 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
Aug 6, 2025
Wan Video 14B t2v
Instagirl V2 is a complete overhaul, trained from the ground up to push the boundaries of realism on Wan 2.2.
Unprecedented Realism: We've massively upgraded the dataset and training method with a focus on photorealistic skin textures, natural lighting, and flawless environmental details.
Greater Diversity: V2 is trained on a much wider and more inclusive dataset, enabling a greater variety of faces, ethnicities, and styles right out of the box.
Better Composition: The model now has a deeper understanding of world composition, resulting in more coherent and believable scenes with fewer artifacts.
This model is trained on WAN 2.2, meaning there are two versions; a high and low noise variant.
Show more

1.6K0 1 2 3 4 5 6 7 8 9.0 1 2 3 4 5 6 7 8 9K
3.5K0 1 2 3 4 5 6 7 8 9.0 1 2 3 4 5 6 7 8 9K
60.4K0 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9.0 1 2 3 4 5 6 7 8 9K
License:
Apache 2.0How to Use
This is a style LoRA, designed to be the base layer of your LoRA stack. It creates the foundational aesthetic of realism, upon which you can add character or concept LoRAs.
Note: The included ZIP archive contains both the high-noise and low-noise LoRA variants, along with our recommended ComfyUI workflows.
Trigger Word:
InstacamRecommended Strength:
1.0. Start here and adjust in small increments.
I would also like to thank Danrisi, who originally taught us how to train LoRAs and helped make our work possible.
