Update Cache-DiT release notes and architecture image

Removed outdated Cache-DiT v1.2.0 release note and added v1.2.1 details. Updated image format for architecture representation.

Author: DefTruth

README.md

@@ -35,11 +35,9 @@
 ## 📖 News 🔥🔥
 <div id="news"></div>
 
-- [2026-01-16]: **[🤗Cache-DiT v1.2.0 Major Release](https://github.com/vipshop/cache-dit)** is ready! 🤗Cache-DiT is A PyTorch-native Inference Engine with Hybrid Cache Acceleration and Parallelism for DiTs. Feel free to take a try!
+- [2026/02] Cache-DiT **[🎉v1.2.1](https://github.com/vipshop/cache-dit)** release is ready, the major updates including: [Ring](https://cache-dit.readthedocs.io/en/latest/user_guide/CONTEXT_PARALLEL) Attention w/ [batched P2P](https://cache-dit.readthedocs.io/en/latest/user_guide/CONTEXT_PARALLEL), [USP](https://cache-dit.readthedocs.io/en/latest/user_guide/CONTEXT_PARALLEL/) (Hybrid Ring and Ulysses), Hybrid 2D and 3D Parallelism (💥[USP + TP](https://cache-dit.readthedocs.io/en/latest/user_guide/HYBRID_PARALLEL/)),  VAE-P Comm overhead reduce.
 
-<div align='center'>
-  <img src=https://github.com/vipshop/cache-dit/raw/main/assets/cache-dit-logo-v2.png height=320px>
-</div>
+![arch](https://github.com/vipshop/cache-dit/raw/main/assets/arch.png)
 
 - Most of my time now is focused on **LLM/VLM** Inference. Please check 📖[Awesome-LLM-Inference](https://github.com/xlite-dev/Awesome-LLM-Inference)  ![](https://img.shields.io/github/stars/xlite-dev/Awesome-LLM-Inference.svg?style=social) and 📖[LeetCUDA](https://github.com/xlite-dev/LeetCUDA)  ![](https://img.shields.io/github/stars/xlite-dev/LeetCUDA.svg?style=social) for more details. Now, [lite.ai.toolkit](https://github.com/xlite-dev/lite.ai.toolkit) ![](https://img.shields.io/github/stars/xlite-dev/lite.ai.toolkit.svg?style=social) is mainly maintained by 🎉[@wangzijian1010](https://github.com/wangzijian1010).