Instructions to use OpenMOSS-Team/MOSS-Image-Base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use OpenMOSS-Team/MOSS-Image-Base with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("OpenMOSS-Team/MOSS-Image-Base", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
- DiffusionBee
wish list
⢠FP32 release of this model, including an FP32 text encoder
⢠This is the big one: You said you plan to release an anime-focused version of this model. Once you have done that please strip all of the anime/illustration out of this model and release a realism-focused update. Since the anime-focused model will be good at doing that this will free up more room in this small model for more realism. You might want to call the anime-focused model MOSS-A and the realism-focused model MOSS-R.
Thank you.
Also, I haven't tested MOSS enough to know yet but am guessing based on my testing of other models. It would be great to have a model that can do authentic text that doesn't look like a font, like a word scrawled onto a surface with various materials. That would be a big advantage to my proposed MOSS-R â to be able to have things written on surfaces and not have it look artificial.