ShoukanLabs

🧪 ShoukanLabs ー召喚Labsー

Shoukan 召喚【しょうかん】 ー can be translated into either Summon or Summoning.

🔎 What is ShoukanLabs?

We are a tiny group of AI enthusiasts founded by ButterCream, and currently comprised of:

• ButterCream
• AshtakaOOf
• Alex
• High Speed Rail Enjoyer
• Neggles
• ZeroCool

We are working on AI Image generation and other generative technologies!

📚 Our Projects

• Future Projects

  • OpenNiji-V3 (placeholder name | Most likely to be built on Muse)

• Current Projects

  • A Large collection of models surrounding TTS (Vokan-V2 included)
  • Muse - A new architecture for efficient text2image generation, based on the Google Paper
  • OpenNiji-Dataset - The dataset that will be used for OpenNiji-V3
    • Dataset preview - See what images the dataset contains

• Previous Projects

  • Vokan - An expressive StyleTTS2 finetune with better 0-shot capabilities
  • OpenNiji - A finetune aimed at replicating Nijijourney on Stable Diffusion.
  • OpenNiji-V2 - A second finetune made to replicate the Nijijourney style more accurately.

Announcements

• We are happy to announce a partnership with Sygil.dev! We're working on some massive things together and hope to share them with you soon!