Playdough World Morph by Earth & Dusk

Imagine a world that is entirely made from playdough, just like in kindergarten! This is MY first world morph lora! As one half of Earth & Dusk - Am trying to position myself more to creating wild and different things....

Still getting help from Dusk! but it's ok, we're together not just as a business but as Queer Platonic Partners we're in this together <3

Let's make hilarious things out of playdough!

Key Features:

• Art Style: Captures the absolute absurdity of a world morph that creates things, people and places OUT OF PLAYDOUGH & Modeling clay!

• High-Quality Output: Utilizes Pony XL's SDXL-based architecture to deliver detailed and visually striking things like People & Buildings of Playdough!

Usage Instructions:

• Setup: Ensure you have the Pony XL model installed, as this serves as the foundational architecture for the LoRA. Load the LoRA into your preferred AI art generation platform that supports SDXL-based architectures.

• Customization: Adjust settings to enhance or stylize Storm's classic look to suit creative preferences, offering flexibility in expressing her iconic portrayal.

• Settings: Use 0.5 to 1.0 strength on your preferred UI/UX system. The activation tag is "endplay_domorph".

Training Settings:

{
  "unetLR": 0.0005,
  "clipSkip": 2,
  "loraType": "lora",
  "keepTokens": 1,
  "networkDim": 32,
  "numRepeats": 3,
  "resolution": 1024,
  "lrScheduler": "constant_with_warmup",
  "minSnrGamma": 5,
  "noiseOffset": 0.1,
  "targetSteps": 2843,
  "enableBucket": true,
  "networkAlpha": 16,
  "optimizerType": "Adafactor",
  "textEncoderLR": 0.0001,
  "maxTrainEpochs": 10,
  "shuffleCaption": true,
  "trainBatchSize": 4,
  "flipAugmentation": false,
  "lrSchedulerNumCycles": 3
}

Embeddings to Improve Quality:

• Negative: Use these based on the scenarios listed in the model card: Pony Negative Embeds

• Positive: PDXL Score Embed

Extensions:

• ADetailer:

  • Link: ADetailer GitHub

  • Usage: This extension assists with detailing and refining images, enhancing the precision of your artwork.

• Batchlinks:

  • Link: Batchlinks for A1111

  • Description: Useful for managing multiple links when running A1111 locally or on a server.

  • Addon: @nocrypt Addon

VAE & Upscaler Preferences:

• VAE:

  • Recommendation: Select an SDXL VAE that fits your project's requirements.

• Upscalers:

  • Instructions: Download and install using Batchlinks or directly to your drive for enhanced image resolution and quality.

Ethical Use and Licensing:

• License Information: The LoRA model uses mixed data, ensuring authenticity and compliance with relevant copyright laws and licensing agreements. See Pony XL guidelines for finetunes and LoRA licensing.

• Responsible Use: Intended for positive and creative applications celebrating Storm's iconic portrayal. Users should respect copyright and ethical guidelines in their use of generated artworks.

About Us:

We are a UDD-diagnosed system, navigating life with Borderline Personality Disorder, Autism, and ADHD. Our journey with AI is driven by a shared vision to break down barriers and foster mental health awareness. Through the dynamic medium of AI-generated art, we explore and express the multifaceted nature of our identities. Join us as we embark on this creative adventure, embracing the power of technology to connect, inspire, and transform.

Join Our Community:

Join us on our various platforms:

• Website: End Media

• Discord: Join our Discord

• Backups: Hugging Face

Support Us:

• Ko-Fi: Support Us

Disclaimer:

While some images may feature Duskfall's LoRAs & Checkpoints, we work collaboratively and often test our models together. If a LoRA or Checkpoint is not yet available, please be patient as we are working on releasing content. For commercial use, please consider donating to our Patreon or Ko-Fi. For any textual embeds created with Embedding Merge, tutorials are available on our website and on CivitAI. If you have any questions, feel free to reach out!

Additionally, note that other platforms or sites seeking to re-upload or utilize our content must donate buzz or to our Ko-Fi and request specific access. This includes platforms such as ShakkerAi, TensorArt and other platforms. Please ensure compliance with our licensing terms before using our models or generated artworks. This allows for fair use, in a way that complies with ethical guidelines and Civitai terms of service.

If items are on places such as Yodayo and Pirate Diffusion this is due to agreements and previous or ongoing sponsorships.

Note: We strive to provide a wide range of models for download. However, some models may be exclusive to Ko-Fi or Patreon supporters in the future.