Text-to-Image
Diffusers
Safetensors
Diffusion Single File
English
Chinese
MOSSImagePipeline
unsloth
fp8
int8
quantized
moss-image
Instructions to use unsloth/MOSS-Image-Base-FP8 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use unsloth/MOSS-Image-Base-FP8 with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("unsloth/MOSS-Image-Base-FP8", dtype=torch.bfloat16, device_map="cuda") prompt = "a photoreal close-up portrait of an older fisherman, weathered skin, sharp catchlights in the eyes, natural window light, 85mm lens (INT8-ConvRot)" image = pipe(prompt).images[0] - Diffusion Single File
How to use unsloth/MOSS-Image-Base-FP8 with Diffusion Single File:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
- DiffusionBee
- Unsloth Desktop
|
Download NOTICE.md from unsloth/MOSS-Image-Base-FP8: direct link, hf CLI and curl.
- Browser
- Download file 624 Bytes
-
https://huggingface.co/unsloth/MOSS-Image-Base-FP8/resolve/main/NOTICE.md
- Command line
-
hf download hf://unsloth/MOSS-Image-Base-FP8/NOTICE.md
-
curl -L -o NOTICE.md https://huggingface.co/unsloth/MOSS-Image-Base-FP8/resolve/main/NOTICE.md
624 Bytes
MOSS-Image-Base is released by OpenMOSS under the Apache License, Version 2.0 (https://huggingface.co/OpenMOSS-Team/MOSS-Image-Base). Copyright the OpenMOSS authors.
Unsloth AI modified this model: the transformer weights were quantized to FP8 and INT8 (plain and ConvRot-rotated). The original weights are available at https://huggingface.co/OpenMOSS-Team/MOSS-Image-Base.
model_index.json, pipeline_moss_image.py and the scheduler/, text_encoder/, tokenizer/, transformer/ and vae/ folders are unmodified copies of https://huggingface.co/OpenMOSS-Team/MOSS-Image-Base at commit 85975793d8b3d432f1b6e27f352abaa4c2a4c0d3.