Improve model card with metadata and links

README.md

@@ -1,3 +1,16 @@
 ---
 license: mit
+library_name: transformers
+pipeline_tag: image-text-to-text
 ---
+
+# PrismaCoder
+
+PrismaCoder is a family of open multimodal code models for generating and editing charts, webpages, diagrams, documents, and animations from text or visual inputs. It is introduced in the paper [PrismaCoder: Pushing the Boundaries of Open Multimodal Code Intelligence through Decomposed Rubric Synthesis](https://huggingface.co/papers/2610.20538).
+
+- **Paper**: [arXiv](https://arxiv.org/abs/2610.20538) | [Hugging Face Paper](https://huggingface.co/papers/2610.20538)
+- **Project page**: [PrismaCoder Homepage](https://qiushisun.github.io/PrismaCoder-Home/)
+- **Code**: [GitHub repository](https://github.com/QiushiSun/PrismaCoder)
+- **Dataset**: [PrismaCoder-2M](https://huggingface.co/datasets/QiushiSun/PrismaCoder-2M)
+
+The model is available in two sizes: **PrismaCoder-9B** and **PrismaCoder-35B-A3B**, with SFT and RL variants.