Add model card and metadata

Hi! I'm Niels from the Hugging Face team. I'm opening this PR to add a model card for the LatentUM model.

This PR adds:
- Metadata including the `any-to-any` pipeline tag and the `apache-2.0` license.
- Links to the research paper and the official GitHub repository.
- A description of the model's key features.
- Sample usage code for image understanding and image generation tasks, based on the official documentation.
- The BibTeX citation for attribution.

This will help users discover and utilize the model more effectively!

README.md

@@ -0,0 +1,89 @@
+---
+pipeline_tag: any-to-any
+license: apache-2.0
+---
+
+# LatentUM
+
+LatentUM is a unified model that represents all modalities within a shared semantic latent space, eliminating the need for pixel-space mediation between visual understanding and generation. This design naturally enables flexible interleaved cross-modal reasoning and generation.
+
+- **Paper:** [LatentUM: Unleashing the Potential of Interleaved Cross-Modal Reasoning via a Latent-Space Unified Model](https://arxiv.org/abs/2604.02097)
+- **Repository:** [https://github.com/SJTU-DENG-Lab/LatentUM](https://github.com/SJTU-DENG-Lab/LatentUM)
+
+## Key Features
+
+- **Shared Semantic Latent Space:** Text and visual tokens share the same space, enabling direct cross-modal reasoning over generated visual content.
+- **MBAQ:** Visual tokenizer trained to preserve VLM understanding behavior rather than pixel reconstruction.
+- **MoME:** Decoupled understanding/generation branches with shared self-attention for cross-modal interaction.
+- **Decoupled Pixel Decoder:** Optional diffusion decoder for pixel rendering, trained independently to keep the latent space semantics-focused.
+
+## Sample Usage
+
+To use this model, first clone the [official repository](https://github.com/SJTU-DENG-Lab/LatentUM) and install the dependencies:
+
+```bash
+git clone https://github.com/SJTU-DENG-Lab/LatentUM.git
+cd LatentUM
+uv sync
+```
+
+### Image Understanding
+
+```python
+import torch
+from model.latentum import LatentUMModel
+
+dtype = torch.bfloat16
+device = "cuda" if torch.cuda.is_available() else "cpu"
+
+model = LatentUMModel.from_pretrained(
+    "SJTU-DENG-Lab/LatentUM-Base",
+    device = device,
+    dtype  = dtype,
+)
+answer = model.answer(
+    "asset/blue_apple.png",
+    "Describe this image.",
+)
+print(answer)
+```
+
+### Image Generation
+
+```python
+import torch
+from model.decoder import LatentUMDecoderModel
+from model.latentum import LatentUMModel
+
+dtype = torch.bfloat16
+device = "cuda" if torch.cuda.is_available() else "cpu"
+
+model = LatentUMModel.from_pretrained(
+    "SJTU-DENG-Lab/LatentUM-Base",
+    device = device,
+    dtype  = dtype,
+)
+decoder = LatentUMDecoderModel.from_pretrained(
+    "SJTU-DENG-Lab/LatentUM-Decoder",
+    device=device,
+    dtype=dtype,
+)
+images = model.generate_images(
+    "a photo of a cute dog",
+    decoder       = decoder,
+    show_progress = True,
+)
+images[0].save("generated.png")
+```
+
+## Citation
+
+```bibtex
+@article{jin2026latentum,
+  title   = {LatentUM: Unleashing the Potential of Interleaved Cross-Modal Reasoning via a Latent-Space Unified Model},
+  author  = {Jiachun Jin and Zetong Zhou and Xiao Yang and Hao Zhang and Pengfei Liu and Jun Zhu and Zhijie Deng},
+  journal = {arXiv preprint arXiv:2604.02097},
+  year    = {2026},
+  url     = {https://arxiv.org/abs/2604.02097}
+}
+```