Zen3 Image SSD

Distilled SDXL-class text-to-image model for faster inference.

Derived from segmind/SSD-1B (Apache-2.0); repackaged by Zen.

Overview

This repository repackages Segmind SSD-1B (a distilled SDXL, StableDiffusionXLPipeline, dual CLIP text encoders). No weights were trained from scratch by Zen.

Developed/repackaged by Hanzo AI and the Zoo Labs Foundation.

Quick Start

from diffusers import StableDiffusionXLPipeline
import torch

pipe = StableDiffusionXLPipeline.from_pretrained("zenlm/zen3-image-ssd", torch_dtype=torch.float16)
pipe = pipe.to("cuda")
image = pipe("A serene mountain landscape at sunset").images[0]
image.save("output.png")

Model Details

License & Attribution

Apache 2.0. This is a derivative repackaging of SSD-1B by Segmind, released under the Apache License 2.0. All rights and attribution for the underlying model belong to the original authors.

NOTICE

This product includes the SSD-1B model developed by Segmind, licensed under the Apache License 2.0.