Image Generation
Collection
A collection of Keynote Technology's Text-to-Image models. • 1 item • Updated
Text-to-Image
Diffusers
Safetensors
English
StableDiffusionPipeline
image-generation
stable-diffusion
import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("Keynote-Technology/TIGMaN-text-to-image", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]
TIGMaN AI - Trained Image Generation Model and Neural AI
TIGMaN is a new cutting-edge image generation model based on the Stable Diffusion 1.5 architecture. TIGMaN is trained exclusively on public domain images, allowing it to be a robust and versatile model while ensuring that it is avaliable for unrestricted use.
Limitations
TIGMaN may face challenges in generating highly detailed or text-reliant images due to the constraints of permissive content datasets. Additionally, TIGMaN may carry biases encountered on the datasets used for training.
Datasets/Sources Used For Training
TIGMaN is trained on images and content from the following sources:
- Pexels: Pexels License (Public Domain)
- LIBRESHOT: CC0 (Public Domain)
- Unsplash: Lite Dataset License (Public Domain)
- opengameart.org: CC0 (Public Domain)
- Authors: CC0 (Public Domain)
- Contributors: CC0 (Public Domain)
- Met Museum Open Access CC0 (Public Domain)
Intended Uses
- Generative art
- Experiments on generative AI
- Research on the limits of the public domain
- Downloads last month
- 14