Instructions to use argmaxinc/mlx-stable-diffusion-3.5-large-4bit-quantized with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- DiffusionKit
How to use argmaxinc/mlx-stable-diffusion-3.5-large-4bit-quantized with DiffusionKit:
# Pipeline for Stable Diffusion 3 from diffusionkit.mlx import DiffusionPipeline pipeline = DiffusionPipeline( shift=3.0, use_t5=False, model_version=argmaxinc/mlx-stable-diffusion-3.5-large-4bit-quantized, low_memory_mode=True, a16=True, w16=True, )
# Image Generation HEIGHT = 512 WIDTH = 512 NUM_STEPS = 50 CFG_WEIGHT = 5 image, _ = pipeline.generate_image( "a photo of a cat", cfg_weight=CFG_WEIGHT, num_steps=NUM_STEPS, latent_size=(HEIGHT // 8, WIDTH // 8), )
- MLX
How to use argmaxinc/mlx-stable-diffusion-3.5-large-4bit-quantized with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir mlx-stable-diffusion-3.5-large-4bit-quantized argmaxinc/mlx-stable-diffusion-3.5-large-4bit-quantized
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: other
|
| 3 |
+
license_name: stabilityai-ai-community
|
| 4 |
+
license_link: >-
|
| 5 |
+
https://huggingface.co/stabilityai/stable-diffusion-3.5-large/blob/main/LICENSE.md
|
| 6 |
+
library_name: diffusionkit
|
| 7 |
+
base_model: stabilityai/stable-diffusion-3.5-large
|
| 8 |
+
tags:
|
| 9 |
+
- text-to-image
|
| 10 |
+
- image-generation
|
| 11 |
+
- mlx
|
| 12 |
+
inference: false
|
| 13 |
+
language:
|
| 14 |
+
- en
|
| 15 |
+
---
|
| 16 |
+
# Stable Diffusion 3 Medium on DiffusionKit MLX!
|
| 17 |
+
## Check out the [original model](https://huggingface.co/stabilityai/stable-diffusion-3.5-large)!
|
| 18 |
+
## Check out the [DiffusionKit](https://github.com/argmaxinc/DiffusionKit) github repository!
|
| 19 |
+

|
| 20 |
+
# Usage
|
| 21 |
+
- ## Create conda environment
|
| 22 |
+
```shell
|
| 23 |
+
conda create -n diffusionkit python=3.11 -y
|
| 24 |
+
conda activate diffusionkit
|
| 25 |
+
pip install diffusionkit
|
| 26 |
+
```
|
| 27 |
+
- ## Run the cli command
|
| 28 |
+
```shell
|
| 29 |
+
diffusionkit-cli --prompt "detailed cinematic dof render of a \
|
| 30 |
+
detailed MacBook Pro on a wooden desk in a dim room with items \
|
| 31 |
+
around, messy dirty room. On the screen are the letters 'SD3 on \
|
| 32 |
+
DiffusionKit' glowing softly. High detail hard surface render" \
|
| 33 |
+
--model-version argmaxinc/mlx-stable-diffusion-3.5-large-4bit-quantized \
|
| 34 |
+
--height 768 \
|
| 35 |
+
--width 1360 \
|
| 36 |
+
--seed 1001 \
|
| 37 |
+
--step 50 \
|
| 38 |
+
--cfg 7 \
|
| 39 |
+
--t5 \
|
| 40 |
+
--output ~/Desktop/sd3_on_mac.png
|
| 41 |
+
```
|