<h1 id="ffusion-core-2023">FFusion Core 2023</h1>⚠️ Experimental — not a polished release.Retrained a personal 2023 SDXL dataset (969 images + original captions) on Zimage Turbo using only Civitai Trainer — no local training, no manual caption editing.<hr /><h3 id="why-this-exists">🧪 Why This Exists</h3>A stress test to see how Zimage Turbo handles heavy repetition — 20 epochs at just 2 repeats on a large abstract dataset. Deliberately overbaked. Results will vary wildly, and that's the point.<hr /><h3 id="what-it-does">🎨 What It Does</h3><ul><li>Abstract backgrounds &amp; gradient compositions</li><li>3D cube/container renders</li><li>Cloud &amp; smoke formations</li><li>Colorful surreal portraits</li><li>Neon-lit geometric shapes</li></ul>Best used as a style layer, background generator, or starting point for abstract designs.<hr /><h3 id="training-config">⚙️ Training Config</h3>Base Model Zimage Turbo Engine ai-toolkit (Civitai Trainer) LoRA rank 32 / alpha 32 Epochs 20 Repeats 2 Resolution 1024 Optimizer AdamW 8-bit LR / Scheduler 1e-4 / constant Text Encoder not trained<hr /><h3 id="how-to-use">💡 How To Use</h3>Trigger word: <code>ffusion</code>WeightExpect 0.3–0.5 Subtle color/style bleed 0.5–0.8 Recommended range 0.6<hr />Dataset and the original 2023 SDXL LoRA available on request.

FFusion Core 2023

<pre><code>{
 "lr": 0.0001,
 "engine": "ai-toolkit",
 "epochs": 20,
 "ecosystem": "zimageturbo",
 "keepTokens": 0,
 "networkDim": 32,
 "numRepeats": 2,
 "resolution": 1024,
 "lrScheduler": "constant",
 "minSnrGamma": null,
 "noiseOffset": null,
 "networkAlpha": 32,
 "optimizerType": "adamw8bit",
 "shuffleTokens": false,
 "textEncoderLr": null,
 "flipAugmentation": false,
 "trainTextEncoder": false
}</code></pre>

Files with SHA256: 2126e14a3905ec56febe8254a7af37a8826c4db632dbeeedafcf75ef46f44108