kenshin commited on
Commit ·
4b7cff0
1
Parent(s): c794cb9
Update model card and showcase assets
Browse files- .gitattributes +1 -0
- README.md +39 -7
- demo.png → assets/demo.jpg +2 -2
- assets/demo_p2.jpg +3 -0
- assets/edit_demo.jpg +3 -0
.gitattributes
CHANGED
|
@@ -34,4 +34,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
demo.png filter=lfs diff=lfs merge=lfs -text
|
|
|
|
| 37 |
tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
|
|
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
demo.png filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
assets/*.jpg filter=lfs diff=lfs merge=lfs -text
|
| 38 |
tokenizer/tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
README.md
CHANGED
|
@@ -1,4 +1,17 @@
|
|
| 1 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
|
| 3 |
<p align="center">
|
| 4 |
Welcome to the official repository for LLaDA-Image, a unified model for high-quality image generation and editing.
|
|
@@ -12,7 +25,18 @@
|
|
| 12 |
</p>
|
| 13 |
|
| 14 |
<p align="center">
|
| 15 |
-
<img src="demo.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
</p>
|
| 17 |
|
| 18 |
## Introduction
|
|
@@ -23,7 +47,7 @@ This repository provides the checkpoints and Diffusers-based inference code for
|
|
| 23 |
|
| 24 |
## News
|
| 25 |
|
| 26 |
-
- **
|
| 27 |
|
| 28 |
## Highlights
|
| 29 |
|
|
@@ -35,10 +59,15 @@ This repository provides the checkpoints and Diffusers-based inference code for
|
|
| 35 |
|
| 36 |
## Model Zoo
|
| 37 |
|
| 38 |
-
| Model
|
| 39 |
-
| --- | --- | ---: | --- |
|
| 40 |
-
| **LLaDA-Image**
|
| 41 |
-
| **LLaDA-Image-Turbo** | Distilled model for fast generation and editing.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 42 |
|
| 43 |
## Quick Start
|
| 44 |
|
|
@@ -50,6 +79,9 @@ The implementation has been used with Python 3.11, PyTorch 2.8, Transformers 4.5
|
|
| 50 |
git clone https://github.com/inclusionAI/LLaDA-Image.git
|
| 51 |
cd LLaDA-Image
|
| 52 |
|
|
|
|
|
|
|
|
|
|
| 53 |
pip install -r requirements.txt
|
| 54 |
```
|
| 55 |
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- en
|
| 4 |
+
- zh
|
| 5 |
+
library_name: diffusers
|
| 6 |
+
pipeline_tag: text-to-image
|
| 7 |
+
tags:
|
| 8 |
+
- image-generation
|
| 9 |
+
- image-editing
|
| 10 |
+
- image-to-image
|
| 11 |
+
- text-to-image
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<h1 align="center">LLaDA-Image: Building Strong Image Generators with Fully Open Training Recipes</h1>
|
| 15 |
|
| 16 |
<p align="center">
|
| 17 |
Welcome to the official repository for LLaDA-Image, a unified model for high-quality image generation and editing.
|
|
|
|
| 25 |
</p>
|
| 26 |
|
| 27 |
<p align="center">
|
| 28 |
+
<img class="not-prose" src="./assets/demo.jpg" alt="LLaDA-Image realistic image generation showcase" width="100%">
|
| 29 |
+
<em>Photorealistic image generation with natural lighting, lifelike details, and coherent scenes.</em>
|
| 30 |
+
</p>
|
| 31 |
+
|
| 32 |
+
<p align="center">
|
| 33 |
+
<img class="not-prose" src="./assets/demo_p2.jpg" alt="LLaDA-Image text rendering and poster generation showcase" width="100%">
|
| 34 |
+
<em>High-quality text rendering and creative poster generation across diverse visual styles.</em>
|
| 35 |
+
</p>
|
| 36 |
+
|
| 37 |
+
<p align="center">
|
| 38 |
+
<img class="not-prose" src="./assets/edit_demo.jpg" alt="LLaDA-Image editing showcase" width="100%">
|
| 39 |
+
<em>Instruction-guided image editing with faithful content preservation and precise visual changes.</em>
|
| 40 |
</p>
|
| 41 |
|
| 42 |
## Introduction
|
|
|
|
| 47 |
|
| 48 |
## News
|
| 49 |
|
| 50 |
+
- **2026-09-0X:** We released the LLaDA-Image Base and Turbo checkpoints together with the inference code.
|
| 51 |
|
| 52 |
## Highlights
|
| 53 |
|
|
|
|
| 59 |
|
| 60 |
## Model Zoo
|
| 61 |
|
| 62 |
+
| Model | Description | Sampling steps | Hugging Face |
|
| 63 |
+
| --------------------- | ------------------------------------------------------------------------------------- | -------------: | ------------------------------------------------------------------------------------- |
|
| 64 |
+
| **LLaDA-Image** | Base model for high-fidelity text-to-image generation and instruction-guided editing. | 50 | [inclusionAI/LLaDA-Image](https://huggingface.co/inclusionAI/LLaDA-Image) |
|
| 65 |
+
| **LLaDA-Image-Turbo** | Distilled model for fast generation and editing. | 4 | [inclusionAI/LLaDA-Image-Turbo](https://huggingface.co/inclusionAI/LLaDA-Image-Turbo) |
|
| 66 |
+
|
| 67 |
+
## Opensource Plan
|
| 68 |
+
|
| 69 |
+
- [x] Inference code and model weights
|
| 70 |
+
- [ ] Training code (coming soon)
|
| 71 |
|
| 72 |
## Quick Start
|
| 73 |
|
|
|
|
| 79 |
git clone https://github.com/inclusionAI/LLaDA-Image.git
|
| 80 |
cd LLaDA-Image
|
| 81 |
|
| 82 |
+
conda create -n llada-image python=3.11 -y
|
| 83 |
+
conda activate llada-image
|
| 84 |
+
|
| 85 |
pip install -r requirements.txt
|
| 86 |
```
|
| 87 |
|
demo.png → assets/demo.jpg
RENAMED
|
File without changes
|
assets/demo_p2.jpg
ADDED
|
Git LFS Details
|
assets/edit_demo.jpg
ADDED
|
Git LFS Details
|