Update README.md
Browse files
README.md
CHANGED
|
@@ -2,9 +2,9 @@
|
|
| 2 |
<!-- markdownlint-disable html -->
|
| 3 |
<!-- markdownlint-disable no-duplicate-header -->
|
| 4 |
|
| 5 |
-
#
|
| 6 |
<div style="display: flex; justify-content: center; gap: 10px;">
|
| 7 |
-
<a href="https://github.com/SkyworkAI/
|
| 8 |
<img src="https://img.shields.io/badge/GitHub-100000?style=flat&logo=github&logoColor=white" alt="GitHub">
|
| 9 |
</a>
|
| 10 |
<a href="#todo">
|
|
@@ -13,13 +13,13 @@
|
|
| 13 |
</div>
|
| 14 |
|
| 15 |
## π Overview
|
| 16 |
-
**
|
| 17 |
|
| 18 |
You can find more visualizations on our [website](#).
|
| 19 |
|
| 20 |
## π₯ Latest Updates
|
| 21 |
|
| 22 |
-
* [2025-05] π Initial release of
|
| 23 |
|
| 24 |
## π Performance Comparison
|
| 25 |
### GameWorld Score Benchmark Comparison
|
|
@@ -120,15 +120,15 @@ You can find more visualizations on our [website](#).
|
|
| 120 |
</table>
|
| 121 |
|
| 122 |
> Double-blind human evaluation by two independent groups across four key dimensions: **Overall Quality**, **Controllability**, **Visual Quality**, and **Temporal Consistency**.
|
| 123 |
-
> Scores represent the percentage of pairwise comparisons in which each method was preferred.
|
| 124 |
|
| 125 |
|
| 126 |
## π οΈ Installation
|
| 127 |
|
| 128 |
1. Clone the repository:
|
| 129 |
```bash
|
| 130 |
-
git clone https://github.com/SkyworkAI/
|
| 131 |
-
cd
|
| 132 |
```
|
| 133 |
|
| 134 |
2. Install dependencies:
|
|
|
|
| 2 |
<!-- markdownlint-disable html -->
|
| 3 |
<!-- markdownlint-disable no-duplicate-header -->
|
| 4 |
|
| 5 |
+
# Matrix-Game-V1: Interactive World Foundation Model
|
| 6 |
<div style="display: flex; justify-content: center; gap: 10px;">
|
| 7 |
+
<a href="https://github.com/SkyworkAI/Matrix-Game-V1">
|
| 8 |
<img src="https://img.shields.io/badge/GitHub-100000?style=flat&logo=github&logoColor=white" alt="GitHub">
|
| 9 |
</a>
|
| 10 |
<a href="#todo">
|
|
|
|
| 13 |
</div>
|
| 14 |
|
| 15 |
## π Overview
|
| 16 |
+
**Matrix-Game** is a 17B-parameter Diffusion Transformer for generating high-resolution, physics-consistent videos in interactive game environments. Trained on large-scale data from Minecraft and Unreal Engine, it understands game physics like collisions, destruction, and item placement. Matrix-Game supports real-time, action-conditioned generation, adapting video content dynamically to user input.
|
| 17 |
|
| 18 |
You can find more visualizations on our [website](#).
|
| 19 |
|
| 20 |
## π₯ Latest Updates
|
| 21 |
|
| 22 |
+
* [2025-05] π Initial release of Matrix-Game-V1
|
| 23 |
|
| 24 |
## π Performance Comparison
|
| 25 |
### GameWorld Score Benchmark Comparison
|
|
|
|
| 120 |
</table>
|
| 121 |
|
| 122 |
> Double-blind human evaluation by two independent groups across four key dimensions: **Overall Quality**, **Controllability**, **Visual Quality**, and **Temporal Consistency**.
|
| 123 |
+
> Scores represent the percentage of pairwise comparisons in which each method was preferred. Matrix-Game consistently outperforms prior models across all metrics and both groups.
|
| 124 |
|
| 125 |
|
| 126 |
## π οΈ Installation
|
| 127 |
|
| 128 |
1. Clone the repository:
|
| 129 |
```bash
|
| 130 |
+
git clone https://github.com/SkyworkAI/Matrix-Game-V1.git
|
| 131 |
+
cd Matrix-Game-V1
|
| 132 |
```
|
| 133 |
|
| 134 |
2. Install dependencies:
|