Update README.md
Browse files
README.md
CHANGED
|
@@ -2,7 +2,6 @@
|
|
| 2 |
|
| 3 |
|
| 4 |
# MammothModa2: A Unified AR-Diffusion Framework for Multimodal Understanding and Generation
|
| 5 |
-
<img src='./doc/logo.png' alt="MammothModa Logo" width="100" style="max-width: 100px; height: auto;">
|
| 6 |
|
| 7 |
[](https://github.com/bytedance/mammothmoda)
|
| 8 |
[](https://mammothmoda2.github.io/)
|
|
@@ -23,15 +22,6 @@ MammothModa2 (Mammoth2) is a unified autoregressive-diffusion (AR-Diffusion) fra
|
|
| 23 |
|
| 24 |
Trained with roughly 60M supervised generation samples and no reliance on pre-trained generators, Mammoth2 demonstrates that a carefully coupled AR-Diffusion architecture can provide high-fidelity generation and editing while maintaining strong multimodal comprehension within a single, parameter- and data-efficient model.
|
| 25 |
|
| 26 |
-
## Show cases
|
| 27 |
-
<!-- <div align="center">
|
| 28 |
-
<img src='./mammoth.png' alt="MammothModa Overview" width="80%">
|
| 29 |
-
</div> -->
|
| 30 |
-
|
| 31 |
-
<div align="center">
|
| 32 |
-
<img src='./doc/mammoth.png' alt="MammothModa2 Show cases" style="max-width: 80%; height: auto;">
|
| 33 |
-
</div>
|
| 34 |
-
|
| 35 |
## 🎉 News
|
| 36 |
- 2025-12-10: 🔥MammothModa2-Dev models are now available at [HuggingFace](https://huggingface.co/bytedance-research/MammothModa2-Dev).
|
| 37 |
- 2025-10-01: 🔥MammothModa2-Preview models are now available at [HuggingFace](https://huggingface.co/bytedance-research/MammothModa2-Preview). **Note: To use the Preview version, please switch to the `qwen25vl` branch.**
|
|
|
|
| 2 |
|
| 3 |
|
| 4 |
# MammothModa2: A Unified AR-Diffusion Framework for Multimodal Understanding and Generation
|
|
|
|
| 5 |
|
| 6 |
[](https://github.com/bytedance/mammothmoda)
|
| 7 |
[](https://mammothmoda2.github.io/)
|
|
|
|
| 22 |
|
| 23 |
Trained with roughly 60M supervised generation samples and no reliance on pre-trained generators, Mammoth2 demonstrates that a carefully coupled AR-Diffusion architecture can provide high-fidelity generation and editing while maintaining strong multimodal comprehension within a single, parameter- and data-efficient model.
|
| 24 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 25 |
## 🎉 News
|
| 26 |
- 2025-12-10: 🔥MammothModa2-Dev models are now available at [HuggingFace](https://huggingface.co/bytedance-research/MammothModa2-Dev).
|
| 27 |
- 2025-10-01: 🔥MammothModa2-Preview models are now available at [HuggingFace](https://huggingface.co/bytedance-research/MammothModa2-Preview). **Note: To use the Preview version, please switch to the `qwen25vl` branch.**
|