<h3 id="heading-69">GAG = Game Asset Generator 📅 Date: <code>2023-10-23T12:19:42</code></h3><h2 id="heading-70">FF_Potion_Generator 🧪⚗️✨</h2><ul><li>Resolution: <code>1024x1024</code></li><li>Architecture: <code>stable-diffusion-xl-v1-base/lora</code></li><li>Network Dim/Rank: <code>32.0</code></li><li>Alpha: <code>1.0</code></li><li>Module: <code>networks.lora</code></li><li>Learning Rates:<ul><li>Overall: <code>0.0005</code></li><li>UNet: <code>0.0005</code></li><li>Text Encoder: <code>5e-05</code></li></ul></li><li>Optimizer: <code>bitsandbytes.optim.adamw.AdamW8bit(weight_decay=0.1)</code></li><li>Scheduler: <code>cosine_with_restarts</code></li><li>Warmup steps: <code>0</code></li><li>Training Details:<ul><li>Epoch: <code>15</code></li><li>Batches per epoch: <code>250</code></li><li>Gradient accumulation steps: <code>1</code></li></ul></li><li>Images:<ul><li>Train images: <code>1000</code></li><li>Regularization images: <code>0</code></li></ul></li><li>Noise &amp; Regularization:<ul><li>Multires noise iterations: <code>6.0</code></li><li>Multires noise discount: <code>0.3</code></li><li>Min SNR gamma: <code>5.0</code></li><li>Zero terminal SNR: <code>True</code></li><li>Max grad norm: <code>1.0</code></li><li>Clip skip: <code>1</code></li></ul></li><li>Dataset: <code>1 Directory</code> with <code>500 images</code></li><li>UNet Weights:<ul><li>Average magnitude: <code>2.9637</code></li><li>Average strength: <code>0.0109</code></li></ul></li><li>Text Encoder Weights:<ul><li>(1) Magnitude: <code>1.7253</code> | Strength: <code>0.0086</code></li><li>(2) Magnitude: <code>1.7251</code> | Strength: <code>0.0067</code></li></ul></li></ul><h3 id="heading-71">🔖 Tags:</h3><pre><code>
potion, painting of one health potion, potion of healing, health potion, magic potions, magical potions, alchemy concept, hyper realistic poison bottle, rendered art, making a potion, potions, fantasy game spell symbol, rpg game item, alchemist bottles, rpg item render, stylized game art, game icon stylized, hyperdetailed scp artifact jar, 3 d icon for mobile game, 3 d render stylized, casting a spell on a potion ... and 904 more!</code></pre> Starting my GAG Serries 🤟 🥃 This one is a Potion LORA for 2.1 SD models. Depending on the model you will get quite different results. <img src="https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/659fb55b-3c4a-49fa-123a-51b622642300/width=525/659fb55b-3c4a-49fa-123a-51b622642300" /><img src="https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/1849160e-202f-4026-e604-043e35305900/width=525/1849160e-202f-4026-e604-043e35305900" /> You can use a background remover while generating and get: <img src="https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/848c54a8-df37-4ff7-8da0-8eb71580d300/width=525/848c54a8-df37-4ff7-8da0-8eb71580d300" /><h3 id="heading-72"> </h3><img src="https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/3e4b1470-9975-4db9-7aff-73f855385600/width=525/3e4b1470-9975-4db9-7aff-73f855385600" /> again this is a test LORA and is heavily dependent on the input model. LyCORIS version will be up for testing in a few hours. And a 2.1 Model is expected in 48-60 hours For PIXALISATION try: <a target="_blank" rel="ugc" href="https://civarchive.com/models/21276/pixhell"> PIXHELL by R6SPY </a>🤟 PS: so far you can use it for demonstration purposes (preview version) with version 1.0 above (lora and models of GAG series) two versions will be available: Licensed - can be licensed for in-game usage upon request (using only my work and licensed images from my end) Community - using sourced free images and sources but you are responsible is used commercially in your project ¯\_(ツ)_/¯ 🤟 🥃

GAG - RPG Potions  |  LoRa XL

Conducting an initial test with the Civitai LoRA trainer on a limited batch. Date: 2023-10-23T12:19:42 Title: FF_Potion_GeneratorResolution: 1024x1024 Architecture: stable-diffusion-xl-v1-base/loraNetwork Dim/Rank: 32.0 Alpha: 1.0Module: networks.loraLearning Rate (LR): 0.0005 UNet LR: 0.0005 TE LR: 5e-05Optimizer: bitsandbytes.optim.adamw.AdamW8bit(weight_decay=0.1)Scheduler: cosine_with_restarts Warmup steps: 0Epoch: 15 Batches per epoch: 250 Gradient accumulation steps: 1Train images: 1000 Regularization images: 0Multires noise iterations: 6.0 Multires noise discount: 0.3Min SNR gamma: 5.0 Zero terminal SNR: True Max grad norm: 1.0 Clip skip: 1Dataset dirs: 1 [img] 500 imagesUNet weight average magnitude: 2.9637495446847635UNet weight average strength: 0.010910434567277522Text Encoder (1) weight average magnitude: 1.7253436262339745Text Encoder (1) weight average strength: 0.0086376313085263Text Encoder (2) weight average magnitude: 1.7251399575500794Text Encoder (2) weight average strength: 0.006686976993287057

FF_Potion_Generator.safetensors

v0.1 LoRA XL CivitAI ver

v 0.1

<h3 id="heading-69">GAG = Game Asset Generator 📅 Date: <code>2023-10-23T12:19:42</code></h3><h2 id="heading-70">FF_Potion_Generator 🧪⚗️✨</h2><ul><li>Resolution: <code>1024x1024</code></li><li>Architecture: <code>stable-diffusion-xl-v1-base/lora</code></li><li>Network Dim/Rank: <code>32.0</code></li><li>Alpha: <code>1.0</code></li><li>Module: <code>networks.lora</code></li><li>Learning Rates:<ul><li>Overall: <code>0.0005</code></li><li>UNet: <code>0.0005</code></li><li>Text Encoder: <code>5e-05</code></li></ul></li><li>Optimizer: <code>bitsandbytes.optim.adamw.AdamW8bit(weight_decay=0.1)</code></li><li>Scheduler: <code>cosine_with_restarts</code></li><li>Warmup steps: <code>0</code></li><li>Training Details:<ul><li>Epoch: <code>15</code></li><li>Batches per epoch: <code>250</code></li><li>Gradient accumulation steps: <code>1</code></li></ul></li><li>Images:<ul><li>Train images: <code>1000</code></li><li>Regularization images: <code>0</code></li></ul></li><li>Noise &amp; Regularization:<ul><li>Multires noise iterations: <code>6.0</code></li><li>Multires noise discount: <code>0.3</code></li><li>Min SNR gamma: <code>5.0</code></li><li>Zero terminal SNR: <code>True</code></li><li>Max grad norm: <code>1.0</code></li><li>Clip skip: <code>1</code></li></ul></li><li>Dataset: <code>1 Directory</code> with <code>500 images</code></li><li>UNet Weights:<ul><li>Average magnitude: <code>2.9637</code></li><li>Average strength: <code>0.0109</code></li></ul></li><li>Text Encoder Weights:<ul><li>(1) Magnitude: <code>1.7253</code> | Strength: <code>0.0086</code></li><li>(2) Magnitude: <code>1.7251</code> | Strength: <code>0.0067</code></li></ul></li></ul><h3 id="heading-71">🔖 Tags:</h3><pre><code>
potion, painting of one health potion, potion of healing, health potion, magic potions, magical potions, alchemy concept, hyper realistic poison bottle, rendered art, making a potion, potions, fantasy game spell symbol, rpg game item, alchemist bottles, rpg item render, stylized game art, game icon stylized, hyperdetailed scp artifact jar, 3 d icon for mobile game, 3 d render stylized, casting a spell on a potion ... and 904 more!</code></pre> Starting my GAG Serries 🤟 🥃 This one is a Potion LORA for 2.1 SD models. Depending on the model you will get quite different results. <img src="https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/659fb55b-3c4a-49fa-123a-51b622642300/width=525/659fb55b-3c4a-49fa-123a-51b622642300" /><img src="https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/1849160e-202f-4026-e604-043e35305900/width=525/1849160e-202f-4026-e604-043e35305900" /> You can use a background remover while generating and get: <img src="https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/848c54a8-df37-4ff7-8da0-8eb71580d300/width=525/848c54a8-df37-4ff7-8da0-8eb71580d300" /><h3 id="heading-72"> </h3><img src="https://imagecache.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/3e4b1470-9975-4db9-7aff-73f855385600/width=525/3e4b1470-9975-4db9-7aff-73f855385600" /> again this is a test LORA and is heavily dependent on the input model. LyCORIS version will be up for testing in a few hours. And a 2.1 Model is expected in 48-60 hours For PIXALISATION try: <a target="_blank" rel="ugc" href="https://civitai.com/models/21276/pixhell"> PIXHELL by R6SPY </a>🤟 PS: so far you can use it for demonstration purposes (preview version) with version 1.0 above (lora and models of GAG series) two versions will be available: Licensed - can be licensed for in-game usage upon request (using only my work and licensed images from my end) Community - using sourced free images and sources but you are responsible is used commercially in your project ¯\_(ツ)_/¯ 🤟 🥃

&lt;p&gt;Conducting an initial test with the Civitai LoRA trainer on a limited batch.&lt;br /&gt;&lt;br /&gt;Date: 2023-10-23T12:19:42 Title: FF_Potion_Generator&lt;/p&gt;&lt;p&gt;Resolution: 1024x1024 Architecture: stable-diffusion-xl-v1-base/lora&lt;/p&gt;&lt;p&gt;Network Dim/Rank: 32.0 Alpha: 1.0&lt;/p&gt;&lt;p&gt;Module: networks.lora&lt;/p&gt;&lt;p&gt;Learning Rate (LR): 0.0005 UNet LR: 0.0005 TE LR: 5e-05&lt;/p&gt;&lt;p&gt;Optimizer: bitsandbytes.optim.adamw.AdamW8bit(weight_decay=0.1)&lt;/p&gt;&lt;p&gt;Scheduler: cosine_with_restarts  Warmup steps: 0&lt;/p&gt;&lt;p&gt;Epoch: 15 Batches per epoch: 250 Gradient accumulation steps: 1&lt;/p&gt;&lt;p&gt;Train images: 1000 Regularization images: 0&lt;/p&gt;&lt;p&gt;Multires noise iterations: 6.0 Multires noise discount: 0.3&lt;/p&gt;&lt;p&gt;Min SNR gamma: 5.0 Zero terminal SNR: True Max grad norm: 1.0  Clip skip: 1&lt;/p&gt;&lt;p&gt;Dataset dirs: 1&lt;/p&gt;&lt;p&gt;        [img] 500 images&lt;/p&gt;&lt;p&gt;UNet weight average magnitude: 2.9637495446847635&lt;/p&gt;&lt;p&gt;UNet weight average strength: 0.010910434567277522&lt;/p&gt;&lt;p&gt;Text Encoder (1) weight average magnitude: 1.7253436262339745&lt;/p&gt;&lt;p&gt;Text Encoder (1) weight average strength: 0.0086376313085263&lt;/p&gt;&lt;p&gt;Text Encoder (2) weight average magnitude: 1.7251399575500794&lt;/p&gt;&lt;p&gt;Text Encoder (2) weight average strength: 0.006686976993287057&lt;/p&gt;

Files with SHA256: e87063dbedb07c0b4f003b12dd928420acb8b9907545f3e2264a0f0b422b57c3