<p>I made some experiments with training parameters on this models for find best options for myself type of dataset.</p><p></p><p><a target="_blank" rel="ugc" href="https://civarchive.com/articles/6466/level-2-comfyui-image-generation-guide">GUIDE HOW TO GET IMAGES LIKE MY PREVIEWS</a></p><p></p><h3 id="collection-of-my-multicharacter-pdxl-loras-e29h3fhqy"><a target="_blank" rel="ugc" href="https://civarchive.com/collections/1684083">Collection of my multicharacter PDXL LoRAs</a></h3><p></p><h2 id="commisions-open-yk0azxf4j">Commisions <a target="_blank" rel="ugc" href="https://ko-fi.com/randomizer89page/commissions">open</a></h2><p></p><h2 id="support-my-work:-d85b5s50b">Support My Work:</h2><p></p><p><a target="_blank" rel="ugc" href="http://ko-fi.com/randomizer89page"><u><img src="https://storage.ko-fi.com/cdn/kofi2.png?v=3" /></u></a></p><p><a target="_blank" rel="ugc" href="https://boosty.to/blessedworld/donate"><span style="color:rgb(34, 139, 230)"><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/adba302a-bf72-4015-9c14-31e41ce9db40/width=525/adba302a-bf72-4015-9c14-31e41ce9db40.jpeg" /></span></a></p><p>Your support helps me create more amazing content and keeps this passion project alive. Every contribution, no matter how small, makes a big difference. Thank you for being awesome!</p><p></p><h2 id="-d3f5ok6o9"><a target="_blank" rel="ugc" href="https://discord.com/invite/5Y8UnmktH5"><u><span style="color:rgb(34, 139, 230)"><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/9ae340ce-8d0f-4e5a-beeb-3c96be090b8a/width=525/9ae340ce-8d0f-4e5a-beeb-3c96be090b8a.jpeg" /></span></u></a></h2>

Kodomo no Jikan | Kokonoe Rin | (Illustrious and PDXL versions)

<p>just for test, and proof i do this test - it merge 50/50 of pony and illustrious lora trained on same datasets, for look how good result be and how compatible it be, my result is:<br />1) quality is medium -&gt; useless to do it;<br />2) i sure it kill flexibility too (even if i don't test it with multiple loras and long prompting i see problems even on small ones)</p><p>why i put it here: for people can look that you can get by this stupid merge idea without made 2 loras. (!it same dataset as on big one!)</p>

Kodomo no Jikan-Kokonoe Rin-ill-pony merge.safetensors

Pony + illustrious merge

Illustrious_XL + outfits

Illustrious_XL

PDXL

<p>I made some experiments with training parameters on this models for find best options for myself type of dataset.</p><p></p><p><a target="_blank" rel="ugc" href="https://civitai.com/articles/6466/level-2-comfyui-image-generation-guide">GUIDE HOW TO GET IMAGES LIKE MY PREVIEWS</a></p><p></p><h3 id="collection-of-my-multicharacter-pdxl-loras-e29h3fhqy"><a target="_blank" rel="ugc" href="https://civitai.com/collections/1684083">Collection of my multicharacter PDXL LoRAs</a></h3><p></p><h2 id="commisions-open-yk0azxf4j">Commisions <a target="_blank" rel="ugc" href="https://ko-fi.com/randomizer89page/commissions">open</a></h2><p></p><h2 id="support-my-work:-d85b5s50b">Support My Work:</h2><p></p><p><a target="_blank" rel="ugc" href="http://ko-fi.com/randomizer89page"><u><img src="https://storage.ko-fi.com/cdn/kofi2.png?v=3" /></u></a></p><p><a target="_blank" rel="ugc" href="https://boosty.to/blessedworld/donate"><span style="color:rgb(34, 139, 230)"><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/adba302a-bf72-4015-9c14-31e41ce9db40/width=525/adba302a-bf72-4015-9c14-31e41ce9db40.jpeg" /></span></a></p><p>Your support helps me create more amazing content and keeps this passion project alive. Every contribution, no matter how small, makes a big difference. Thank you for being awesome!</p><p></p><h2 id="-d3f5ok6o9"><a target="_blank" rel="ugc" href="https://discord.com/invite/5Y8UnmktH5"><u><span style="color:rgb(34, 139, 230)"><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/9ae340ce-8d0f-4e5a-beeb-3c96be090b8a/width=525/9ae340ce-8d0f-4e5a-beeb-3c96be090b8a.jpeg" /></span></u></a></h2>

&lt;p&gt;just for test, and proof i do this test - it merge 50/50 of pony and illustrious lora trained on same datasets, for look how good result be and how compatible it be, my result is:&lt;br /&gt;1) quality is medium -&amp;gt; useless to do it;&lt;br /&gt;2) i sure it kill flexibility too (even if i don&#39;t test it with multiple loras and long prompting i see problems even on small ones)&lt;/p&gt;&lt;p&gt;why i put it here: for people can look that you can get by this stupid merge idea without made 2 loras. (!it same dataset as on big one!)&lt;/p&gt;

Files with SHA256: 237c6b571878f87512b20b7c79ecdd349445b17e3b539a4b4016c57478865a4a