<h1 id="the-best-way-to-support-me-is-to-subscribe-or-write-some-comment-on-myyoutube-where-i-shitpost-almost-everyday"><span style="color:rgb(250, 82, 82)"><strong>The best way to support me is to subscribe or write some comment on my</strong></span><strong><br /></strong><a target="_blank" rel="ugc" href="https://www.youtube.com/channel/UCNw3_Z8z2uVPwpSegK3mYvA"><span style="color:rgb(250, 82, 82)"><strong><em>YouTube - Where i shitpost almost everyday</em></strong></span></a></h1><h2 id="real-description-part:-allborudatav1-contains-every-possible-image-on-danboru-that-has-visible-giant-+-second-smaller-character-that-i-picked-as-good-quality-ones-made-for-a-general-concept-and-experiment-how-this-one-gonna-perform.">REAL DESCRIPTION PART: <br /><br />ALLBoruDataV1 - Contains every possible image on Danboru that has visible Giant + second smaller character that i picked as good quality ones, made for a general Concept and experiment how this one gonna perform.<br /></h2><h2 id="previous-version-were-made-based-on-search-with-size-difference-tag.-huge-insertionimpossible-fit-search-basically-end-up-with-very-similar-concept-but-completely-different-data-set-and-tags.">Previous version were made based on search with Size Difference tag. <br />Huge Insertion/Impossible Fit search basically end up with very similar concept but completely different data set and tags.</h2><p></p><h2 id="huge-insertion:-this-time-search-based-on-huge-insertion-+-stomach-bulge-tags-rather-than-size-difference-so-it-has-completely-separate-data-set-than-previous-versions.">Huge Insertion: This time search based on huge insertion + stomach bulge tags rather than size difference, so it has completely separate data set than previous versions.</h2><h2 id="wan-2.1-dragon-is-t2v-model-trained-on-civit-was-the-experiment-that-seems-failed-for-now-i-can't-continue-wan-loras-since-is-not-possible-on-civit-and-i-don't-have-local-set-up-that-is-even-near-close-to-training-local-loras."><br /><span style="color:rgb(253, 126, 20)">WAN 2.1 Dragon is T2V model trained on Civit was the experiment that seems failed, for now i can't continue WAN loras since is not possible on Civit and i don't have local set up that is even near close to training local loras.</span></h2><h3></h3><p>I saw there are Bigger dom and Human Onahole models, but wanted to try something that will make dom even bigger.</p><p>Most of images i found for it contained 3 groups of creatues:</p><p>Humanoid (i put here evrything that had hands)</p><p>Dragon (obvious, i really wanted to tagg it "bad dragon" but i'm 99% sure that it is alredy functional tag on site)</p><p>Beast/Monster - evrything that was not humanoid or dragon <br /></p><h3 id="changelog:-v3.0-for-il"><span style="color:rgb(64, 192, 87)">Changelog: V3.0 for IL</span></h3><ul><li><p>has much bigger data, V2 got 20 per monster/dragon/humanoid</p></li><li><p>this one has 60 of each,</p></li><li><p>trained on WAINSFW 12 instead of base 0.1 IL</p></li><li><p>Added onahole "pose"</p></li></ul><h3 id="changelog:-v1.0-was-breaking-faces-way-too-often-so-i-tried-once-again."><br /><span style="color:rgb(64, 192, 87)">Changelog: V1.0 was breaking faces way too often so i tried once again.</span></h3><ul><li><p>Added tags: master piece / best quality mostly for stuff from studio fow / D-rex.</p></li><li><p>added tags: bad quality, bad face, low quality face, badsubface, for negative tagging.</p></li><li><p>removed some pictures that i considered bad quality.</p></li><li><p>Trained without beast this time to see how it changes things, just humanoid and dragons.</p></li><li><p>removed the typos</p></li></ul><h1 id="my-socials:discord-server"><span style="color:rgb(130, 201, 30)"><strong><em>My socials</em>:</strong></span><br /><a target="_blank" rel="ugc" href="https://discord.gg/SGvX8dckDE"><span style="color:rgb(76, 110, 245)"><strong><em>DISCORD SERVER</em></strong></span></a></h1><h1 id="if-you-want-to-contact-me-discordcivit-are-best-options-since-i-check-those-everyday!"><span style="color:rgb(76, 110, 245)">If You want to contact me Discord/Civit are best options since i check those everyday!</span></h1><p>My loras are available from my official accounts on:<br />Tensor Art: <a target="_blank" rel="ugc" href="https://tensor.art/u/836328663648595213">https://tensor.art/u/836328663648595213</a><br />CivitAI: <a target="_blank" rel="ugc" href="https://civarchive.com/user/Shrekman17">https://civarchive.com/user/Shrekman17</a><br />Seeart: <a target="_blank" rel="ugc" href="https://www.seaart.ai/user/ab14b528c289f390e228b8c5d6f497f1?u_code=DAS7ETB">https://www.seaart.ai/user/ab14b528c289f390e228b8c5d6f497f1?u_code=DAS7ETB</a></p>

Giant Dom IL/Pony | Fantasy Sex concept

<p>When i made v1 and 1.1 i was still thinking that for training it is "the more the better" now i looked at this data and it was a lot of garbage, reduced it to 30 files to see how it is gonna perform this time</p>

Files with SHA256: 2d3e165c279fe81e504b53b22d3765c078f3c9ffd01bae43a3e182da318c1981