Text-to-Image
Diffusers
Safetensors
MLX
Flux2KleinPipeline
1-bit
apple-silicon
on-device
diffusion
flux
prismml
bonsai
Instructions to use prism-ml/bonsai-image-binary-4B-mlx-1bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use prism-ml/bonsai-image-binary-4B-mlx-1bit with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("prism-ml/bonsai-image-binary-4B-mlx-1bit", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - MLX
How to use prism-ml/bonsai-image-binary-4B-mlx-1bit with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir bonsai-image-binary-4B-mlx-1bit prism-ml/bonsai-image-binary-4B-mlx-1bit
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Draw Things
- DiffusionBee
Hmm, pretty fast on RTX 3090 24 GB VRAM, though why censored?
#1
by JLouisBiz - opened
Why is the model censored? Like using the word "bloody" in nice context with the love heart is rejected.
I think you should have uncensored models.
Runs fast on RTX 3090 24 GB VRAM
Your github instruction wasn't that right. You forgot to say how to activate environment... and few other steps.
no reason ...
Your github instruction wasn't that right. You forgot to say how to activate environment... and few other steps.
what was the missing instructions? could you make github issue or PR?
the filtering is on application level probably.