20 shown - 165 total matches
Note: If you have ANY issues with nodes not downloading, read the notes or reach out. There's nothing that special about any of them that aren't co...
Create
<p><strong>V5.5 Refinement Update:</strong></p><ul><li><p>Improved prompt adherence for clothing details, garment types and color accuracy</p></li><li><p>Better pose interpretation — sitting, standing, reclining and dynamic poses render more consistently</p></li><li><p>Refined hand anatomy with improved finger definition and object interaction</p></li><li><p>Enhanced multi-figure composition and spatial relationships</p></li><li><p>Stronger facial detail retention in full-body and mid-range shots</p></li><li><p>Overall aesthetic polish with maintained realism and lighting quality</p></li></ul><p><strong>What's unchanged:</strong></p><ul><li><p>Same great skin textures, lighting and style range from V5</p></li><li><p>Fully compatible with existing workflows and LoRAs</p></li></ul><p><strong>Recommended Samplers:</strong></p><ul><li><p>DPM SDE , Euler A,</p></li></ul><p><strong>Optimal Settings:</strong></p><ul><li><p>Steps: 27+</p></li><li><p>CFG Scale: 5</p></li><li><p>Clip Skip: 2</p></li><li><p>Resolution: Any SDXL resolution, 896x1152</p></li></ul><p>Join me on <a target="_blank" rel="ugc" href="https://brandulate.com/"><strong>My Patreon</strong></a> for exclusive perks and early access to unique resources.</p>
Source context: 216 downloads · Type Checkpoint · Base model Illustrious
Welcome to my 💫🎦 Friendly LTX-2.3 T2V+I2V+Lipsync ✨ Less mess, more magic Unified Light - Lipsync all-in one version with single-stage video genera...
Create
A unified, modular SDXL/Pony/Illustrious powerhouse designed to consolidate professional-grade tools into a single, cohesive environment. This work...
Create
Online Workflow： https://www.runninghub.cn/post/2035642836963565570/?inviteCode=rh-v1121 https://www.runninghub.cn/post/2035643056212414465/?invite...
Create
<p>Join The Tinkerer on <a target="_blank" rel="ugc" href="https://whop.com/joined/cyberdelia-ai-lab/products/the-tinkerer/">Whop</a> to download this model and many more — all for one monthly payment. Plus, get early releases, private tools <strong>and</strong> a lot more.<br />👉 <a target="_blank" rel="ugc" href="https://whop.com/joined/cyberdelia-ai-lab/products/the-tinkerer/">Join on Whop</a></p><p>V5.0 is a more distinct stylistic departure from the V4.x line, not just a subtle refinement. It's got improved Danbooru tag compatibility and a noticeably different look compared to V4.5 - check the pinned comparison images for a side-by-side.</p><p>V4.0 and V4.5 are still excellent checkpoints and aren't going anywhere. V5.0 is just a different flavor, aimed at users who prefer its particular style or work mostly with Danbooru-style prompting.</p>
Source context: 14 downloads · Type Checkpoint · Base model Pony
<p>I’ve been working on this version for the past three months. While there are still many areas I want to refine without simply blaming "architectural limitations," what I’ve achieved for Version 3 is a massive leap forward.</p><p>As promised, the changes between versions are substantial—you can clearly see the progress in the comparison images. This isn't just another incremental update; it’s a result of deep R&D and surgical weight tuning.</p>
Source context: 1080 downloads · Type Checkpoint · Base model Illustrious
<p>Sampler: heunpp2 + linear_quadratic schedule</p><p>seeds_3 + normal is better sampler.</p><p>CFG Scale: 1 – 2，Steps: 8 – 20</p><p>Recommended Speed: CFG: 1, Step: 10</p><p>Recommended Quality: CFG: 1.5, Step: 20</p><p>Prompting Tip: The more detailed and fluent your prompt, the closer the result matches your vision. Please ensure your prompts are well-structured.</p>
Source context: 289 downloads · Type Checkpoint · Base model ZImageTurbo
Note: If you have ANY issues with nodes not downloading, read the notes or reach out. There's nothing that special about any of them that aren't co...
Create
<p><span><strong>SemiReal Engine</strong> is a high‑precision semi‑realistic SDXL model positioned exactly between photographic realism and premium 2.5D illustration. It delivers clean shapes, smooth textures, precise edges, and a soft stylized depth that makes it ideal for:</span></p><ul><li><p><span>portraits</span></p></li><li><p><span>half‑body shots<...
Source context: 38 downloads · Type Checkpoint · Base model SDXL 1.0
Models Used model = wan2.1_vace_14B_fp16.safetensors (WAN 2.1 Video Diffusion – core generation model) lora = Wan21_CausVid_14B_T2V_lora_rank32.saf...
Create
<p>Realism_EngineV3 is a next‑generation ultra‑realistic rendering engine for SDXL, designed for <strong>close‑up portraits</strong>, <strong>half‑body shots</strong>, <strong>macro details</strong>, <strong>animals</strong>, <strong>flowers</strong>, <strong>plants</strong>, <strong>objects</strong>, and <strong>landscapes</strong>. Its core philosophy: <strong>physical accuracy, micro‑detail fidelity, and total stylistic control</strong>.</p><p>🔥 <strong>Key Strengths</strong></p><p><strong>1. Close‑Up Portraits</strong></p><ul><li><p>enhanced skin micro‑texture</p></li><li><p>realistic pores, peach fuzz, and eye reflections</p></li><li><p>physically accurate natural lighting (golden hour, moonlight, overcast)</p></li><li><p>flawless 85mm lens simulation</p></li></ul><p><strong>2. Half‑Body Shots</strong></p><ul><li><p>improved fabric realism</p></li><li><p>natural folds, stitching, and material reflectance</p></li><li><p>more stable hand rendering</p></li><li><p>clean depth separation</p></li></ul><p><strong>3. Macro Detail Engine</strong></p><p>Exceptional performance on:</p><ul><li><p>animals (fur, feathers, scales)</p></li><li><p>flowers (petal translucency, pollen detail)</p></li><li><p>plants (veins, moisture, surface texture)</p></li><li><p>objects (metal, glass, ceramic, micro‑scratches)</p></li></ul><p><strong>4. Landscapes</strong></p><ul><li><p>improved atmospheric depth</p></li><li><p>natural haze and volumetric scattering</p></li><li><p>realistic color grading</p></li><li><p>physically coherent ambient light</p></li></ul><p>⚙️ <strong>Technical Optimizati...
Source context: 152 downloads · Type Checkpoint · Base model SDXL 1.0
Model repo on Hugging Face Hub.
Source context: 37968104 downloads · 1039 likes · Pipeline image-classification · Library transformers · Repo Falconsai/nsfw_image_detection
Updated 04/06/2026 Note that the examples of the new version are only posted here, Github does NOT have the new examples, the code is updated thoug...
Create
<p>New base model based on JANKU and <a target="_blank" rel="ugc" href="https://civitai.com/models/2167995/chenkin-noob-xl">Chenkin v0.2</a> with training</p><ul><li><p>Initial release to add stability to Chenkin image output.</p></li><li><p>Thicker lines on characters.</p></li><li><p>More detailed backgrounds</p></li><li><p>Compared to previous version of JANKU, there is less artist knowledge.</p></li></ul>
Source context: 9718 downloads · Type Checkpoint · Base model Illustrious
<p>Join The Tinkerer on <a target="_blank" rel="ugc" href="https://whop.com/joined/cyberdelia-ai-lab/products/the-tinkerer/">Whop</a> to download this model and many more — all for one monthly payment. Plus, get early releases, private tools <strong>and</strong> a lot more.<br />👉 <a target="_blank" rel="ugc" href="https://whop.com/joined/cyberdelia-ai-lab/products/the-tinkerer/">Join on Whop</a><br /><br />A solid upgrade with significantly improved NSFW performance, while still delivering beautiful results.</p>
Source context: 12 downloads · Type Checkpoint · Base model SDXL 1.0
<p>A nude/NSFW capable model that focuses on images with a female primary subject.</p><p>It is heavily biased towards nude images.</p><p></p><h2 id="recommended-settings"><span style="color:rgb(250, 176, 5)">Recommended Settings</span></h2><p><span style="color:rgb(64, 192, 87)"><strong>Sampler is dpmpp_2m (DPM++ 2M)</strong></span> and the <span style="c...
Source context: 3 downloads · Type Checkpoint · Base model Flux.1 D
<ul><li><p>Highres training is in progress. Trained for much longer at 1024 resolution than preview2.</p></li><li><p>Expanded dataset to help learn less common artists (roughly 50-100 post count).</p></li></ul>
Source context: 3815 downloads · Type Checkpoint · Base model Anima
<p>Major alternative merge, changes in overall style, lighting and prompt adherence.</p>
Source context: 234 downloads · Type Checkpoint · Base model Illustrious
<ul><li><p><a rel="ugc" href="https://civitai.com/models/2458426?modelVersionId=2836417">Anima preview 3</a> based.</p></li><li><p>Added 1792px resolution dataset for learning. (multi-resolution datasets with 1280px,1536px and 1792px)</p></li></ul>
Source context: 241 downloads · Type Checkpoint · Base model Anima
