<h3 id="aitk-humanoids-beta-version-update">🔥 AITK-Humanoids — BETA Version Update</h3>Here’s a fresh update for those following the project: I’ve just trained a BETA version of AITK-Humanoids, based on a new dataset of 255 images, retrained from scratch.The primary goal of this update was to solve the robotic humanoid cliché issue — and yes, it’s working beautifully. Expect much more variety in alien-inspired humanoids, with better integration of synthetic textures, plating, and facial structure. There’s also a notable boost in overall shape diversity for alien types.⚠️ However, this version is a bit more unstable than I usually go for. Here’s the breakdown:<ul><li>90% of the outputs are solid and usable</li><li>8% are insane — stunning compositions, skin textures, and creative forms</li><li>2% are just broken — visual glitches or anatomical weirdness</li></ul>Why release it? Because the level of creativity and texture quality in this checkpoint was simply too good to ignore. Personally, I’d rather have a few unusable outputs in exchange for occasional masterpieces.✅ Stability Tip Combine this LoRA with:<ul><li><code>AITK-Cine</code> at +0.5</li><li><code>AntiFlux</code> at -0.5</li></ul>With this setup, I’ve had zero unusable outputs so far — and the aesthetic quality stays sky-high.<hr />🧠 A few clarifications and prompt tips:<ul><li>The LoRA has a bit of a mind of its own — it won’t always follow your prompt exactly, and that’s part of its strength. You’ll get far more variation than you would from a base Flux-style LoRA.</li><li>As mentioned in the earlier version: you can go highly descriptive, or just give it space to invent.</li><li>Important: "Alien" ≠ "Humanoid". If your character should be more human in form, use terms like “humanoid” or “female humanoid.” If you want a non-human silhouette or structure, drop the term humanoid and just use “alien.”</li><li>Also, using “woman” can sometimes throw off results — "female" and "male" are more neutral and reliable.</li></ul><hr />🛠️ Known BETA Issues (to be addressed next):<ul><li>Model struggles with rounder or fuller body types — defaults too often to slender or muscular frames</li><li>Obese or heavily built characters are rare</li><li>Robotic humanoids now work great — but I’ll refine even further in the next update</li></ul>If you notice other issues or edge cases, feel free to send feedback — it genuinely helps shape future iterations.<hr /><h3 id="aitk-humanoids-first-release">AITK-Humanoids — First Release</h3>Hey everyone — I’m back at it.After a break from training LoRAs, I decided it was time to tackle one of the weaker points of the base Flux dev model: its inability to properly generate interesting, diverse humanoid characters. That’s exactly what AITK-Humanoids aims to fix.This LoRA allows you to generate a broad spectrum of humanoid forms — from stylized aliens and fantasy subspecies to underrepresented human ethnicities. While I experimented with early hominids like Neanderthals, Australopithecus, and Homo Erectus, those concepts didn’t fully take in this version. Still, feel free to try — edge cases can yield interesting results.<hr /><h3 id="lora-specs-and-dataset">⚙️ LoRA Specs &amp; Dataset</h3><ul><li>Weight: 1.0 is optimal — it holds strong on its own</li><li>Trained on: 99 custom images</li><li>Sources: A mix of real photography and carefully curated artificial data</li><li>Resolution range: 512 → 1536px (I typically generate at 1MP)</li><li>Sampler/Scheduler: IPNDM + Scheduler Beta (20 steps recommended)</li></ul>This version excels at rendering:<ul><li>Diverse humanoid shapes, skin textures, and body types</li><li>Most skin tones, including rarer color combinations</li><li>Both SFW and NSFW content (I don’t gate usage — you do you)</li></ul>One limitation to note: robotic or cybernetic concepts are not well understood yet — that’s planned for a future update.<hr /><h3 id="prompting-tips-and-template">🧠 Prompting Tips &amp; Template</h3>This LoRA was captioned using the following structure:<code>[Angle], [Character &amp; attributes], [optional pose], wearing [Clothes], in [Environment], with [Lighting Type]</code>The more precise you are, the more control you’ll have. But if you're looking for creative variation, stay intentionally vague — the model will surprise you.🔄 Aspect Ratios:All ratios were included in training. It should perform well across vertical, horizontal, and square formats.<hr /><h3 id="use-this-for:">✅ Use This For:</h3><ul><li>Sci-fi or fantasy humanoids</li><li>Ethnic diversity and inclusivity in characters</li><li>Concept art (photoreal or stylized)</li><li>NSFW or artistic nudity, with realistic anatomy and variation</li></ul><hr /><h3 id="known-limitations:">❌ Known Limitations:</h3><ul><li>Not yet reliable for robotic or mecha-inspired humanoids</li><li>Primitive species (Neanderthals, Homo Habilis, etc.) need refinement</li><li>May not play well with other photoreal LoRAs — use solo for best results</li></ul><hr />Feel free to share your results or experiments. I’m always curious to see how you push this LoRA beyond its intended scope. This is just the start — future versions will go deeper into speculative evolution, robotics, and expanded diversity.

Files with SHA256: d46f1ab1307e08191e2c0053e33f72e781a2ba4cc935bec696238b9fac1d1a31