<h3 id="https:ko-fi.complaytime_ai"><a target="_blank" rel="ugc" href="https://ko-fi.com/playtime_ai">https://ko-fi.com/playtime_ai</a></h3><p></p><h3 id="triggers:-a-beautiful-girl-or-a-beautiful-young-woman-or-a-beautiful-woman...-each-will-yield-different-results.">Triggers: <em>a beautiful girl OR a beautiful young woman OR a beautiful woman... Each will yield different results.</em></h3><h1 id="wan-2.2-v4"><strong><u>Wan 2.2 - v4</u></strong></h1><p><strong>v4 was trained with a new dataset and a new training pipeline / training parameters. I think my new workflow is giving better results for Wan 2.2.</strong></p><p></p><h1 id="wan-2.2-v3"><strong><u>Wan 2.2 - v3</u></strong></h1><p>Expanded data set yet again and tweaked a few training settings.</p><h2 id="also-great-for-t2i"><span style="color:rgb(253, 126, 20)"><strong>Also great for T2I</strong></span></h2><h3 id="triggers:-a-beautiful-girl-or-a-beautiful-young-woman-or-a-beautiful-woman...-each-will-yield-different-results.">Triggers: <em>a beautiful girl OR a beautiful young woman OR a beautiful woman... Each will yield different results.</em></h3><p></p><h1 id="v4.1-more-diversity!"><strong><u>v4.1 - More Diversity!</u></strong></h1><h3 id="this-version-was-trained-on-an-expanded-more-ethnically-diverse-dataset.-all-of-the-previous-improvements-still-apply.">This version was trained on an expanded, more ethnically diverse dataset. All of the previous improvements still apply.</h3><p></p><h1 id="v3-now-with-nudity!"><strong><u>v3 - Now with nudity!</u></strong></h1><p><strong>I noticed that the previous version was not wanting to generate nudity. I refined the dataset and added in some nude images. Problem solved!</strong></p><p></p><p><strong>Note: </strong>Using the word<strong> '</strong>woman' vs. 'girl' makes a big difference in the output.</p><p></p><p>Wan v1 was a little over trained. I revised the dataset and reran the training with new LR and settings to create v2. v2 should be more compatible with combining loras.</p><p></p><p>This model was trained on 50 images of 50 different women. All pictures are publicly available. No trigger is necessary. Just prompt as usual and turn on the lora.</p><p></p><p><span style="color:rgb(250, 82, 82)"><strong>(If you have an idea or a request, send me a message. I can afford my own coffee.)</strong></span></p>

Instagram Women (Wan 2.1 & 2.2) - Official 2.2 Release!

T2V - Instagram Women v4,1.safetensors

Instagram Women

T2V-14B - v4.1

v4 Wan2.2 LOW

v4 Wan2.2 HIGH

Qwen

Illustrious

Pony

SDXL

Flux v2

v3 Wan2.2 - LOW - 14B

v3 Wan2.2 - HIGH - 14B

v2 Wan2.2 - T2V - LOW

v2 Wan 2.2 - T2V - HIGH

Wan2.2-T2V-LOW

Wan2.2-T2V-HIGH

depricated- v0.3 14B

depricated - T2V-14B

Flux

Wan2.1 - T2V-14B - v3

Wan2.1 - T2V-14B - v2

hunyuan

Wan - T2V-14B

v4.1 - More Diversity!
This version was trained on an expanded, more ethnically diverse dataset. All of the previous improvements still apply.

v4.1

<p>This LoRA adds Instagram women to Wan. </p>

Files with SHA256: a23062fea6a2a9fc57a6881f76946ad93f62087b1e735a5fdb48d4ec67d07e0e