Explore
laion-ai/erlich
Generate a logo using text.
nightmareai/arf-svox2
Artistic Radiance Fields - Transfer the style of an image to a 3D scene (NeRF)
borisdayma/dalle-mini
Generate images from a text prompt
pixray/text2image
Uses pixray to generate an image from text prompt
google-research/maxim
Multi-Axis MLP for Image Processing
bytedance/piano-transcription
high-resolution piano transcription system: detects piano notes from audio
Collections
Style transfer
Models that take a content image and a style reference to produce a new image
paper11667/clipstyler, ariel415el/gpdm, ptran1203/pytorch-animegan, jiupinjia/stylized-neural-painting-oil, mistake0316/style-transfer-clip...
ML makeovers
Models that let you change facial features
orpatashnik/styleclip, rinongal/stylegan-nada, yuval-alaluf/sam, yuval-alaluf/restyle_encoder, eladrich/pixel2style2pixel...
Image restoration
Models that improve or restore images by deblurring, colorization, and removing noise
tencentarc/gfpgan, jingyunliang/swinir, microsoft/bringing-old-photos-back-to-life, yangxy/gpen, google-research/maxim...
Text to image
Models that generate images from text prompts
pixray/text2image, laion-ai/erlich, mehdidc/feed_forward_vqgan_clip, jack000/glid-3-xl, facebookresearch/ic_gan...
Popular models
nicholascelestin/dalle-mega
Will decommission soon in favor of official dalle-mini replicate model.
afiaka87/laionide-v3
GLIDE finetuned on LAION5B, then more on curated datasets.
jack000/glid-3-xl
A 1.4B parameter text2im model from CompVis, finetuned on CLIP text embeds and curated data.
orpatashnik/styleclip
Text-Driven Manipulation of StyleGAN Imagery
dribnet/pixray-text2image
Uses pixray to generate an image from text prompt
Latest models
storymy/take-off-eyeglasses
Remove eyeglasses and shadows from photo
nicholascelestin/latent-diffusion
High-Resolution Image Synthesis with Latent Diffusion Models
nightmareai/real-esrgan
Real-ESRGAN with optional face correction and adjustable upscale
wzx0826/lbnet
Lightweight Bimodal Network for Single-Image Super-Resolution via Symmetric CNN and Recursive Transformer
thudm/cogview2
Generate vivid images for any (Chinese / English) text
