Improve model card: Add Diffusers library, text-to-image pipeline tag, and HF paper link
Browse filesThis PR enhances the model card for `sanaka87/Show-o-RecA` by:
- Updating the `pipeline_tag` from `any-to-any` to `text-to-image` in the metadata, which better reflects the model's primary capabilities in image generation and editing and improves discoverability on the Hugging Face Hub.
- Adding `library_name: diffusers` to the metadata. The `config.json` file indicates compatibility with Diffusers, enabling the automated "how to use" widget on the model page.
- Including a direct link to the Hugging Face paper page ([Reconstruction Alignment Improves Unified Multimodal Models](https://huggingface.co/papers/2509.07295)) as a new badge in the "Method" section.
These improvements will make the model more accessible and easier for users to understand and integrate.
|
@@ -1,14 +1,16 @@
|
|
| 1 |
---
|
| 2 |
-
|
|
|
|
| 3 |
datasets:
|
| 4 |
- brivangl/midjourney-v6-llava
|
| 5 |
language:
|
| 6 |
- en
|
| 7 |
- zh
|
| 8 |
-
|
| 9 |
-
|
| 10 |
-
|
| 11 |
---
|
|
|
|
| 12 |
# Show-o-RecA
|
| 13 |
|
| 14 |
> A self-supervised training framework that aligns understanding and generation in modest compute, with huge **zero-shot** gain on generation and editing capability.
|
|
@@ -19,6 +21,7 @@ This repository hosts the model weights for **Show-o-RecA**. For installation, u
|
|
| 19 |
|
| 20 |
[](https://arxiv.org/pdf/2509.07295)
|
| 21 |
[](https://arxiv.org/abs/2509.07295)
|
|
|
|
| 22 |
[](https://github.com/HorizonWind2004/reconstruction-alignment)
|
| 23 |
[](https://huggingface.co/collections/sanaka87/realign-68ad2176380355a3dcedc068)
|
| 24 |
[-fcd022?style=for-the-badge&logo=huggingface&logoColor=000)](https://huggingface.co/spaces/sanaka87/BAGEL-ReAlign)
|
|
|
|
| 1 |
---
|
| 2 |
+
base_model:
|
| 3 |
+
- showlab/show-o-w-clip-vit
|
| 4 |
datasets:
|
| 5 |
- brivangl/midjourney-v6-llava
|
| 6 |
language:
|
| 7 |
- en
|
| 8 |
- zh
|
| 9 |
+
license: apache-2.0
|
| 10 |
+
pipeline_tag: text-to-image
|
| 11 |
+
library_name: diffusers
|
| 12 |
---
|
| 13 |
+
|
| 14 |
# Show-o-RecA
|
| 15 |
|
| 16 |
> A self-supervised training framework that aligns understanding and generation in modest compute, with huge **zero-shot** gain on generation and editing capability.
|
|
|
|
| 21 |
|
| 22 |
[](https://arxiv.org/pdf/2509.07295)
|
| 23 |
[](https://arxiv.org/abs/2509.07295)
|
| 24 |
+
[](https://huggingface.co/papers/2509.07295)
|
| 25 |
[](https://github.com/HorizonWind2004/reconstruction-alignment)
|
| 26 |
[](https://huggingface.co/collections/sanaka87/realign-68ad2176380355a3dcedc068)
|
| 27 |
[-fcd022?style=for-the-badge&logo=huggingface&logoColor=000)](https://huggingface.co/spaces/sanaka87/BAGEL-ReAlign)
|