Update README.md

README.md

@@ -11,15 +11,30 @@ pinned: false
     <img src="https://cdn-uploads.huggingface.co/production/uploads/656d9adce8bf55919aca7c3f/j4NWHo9opdcv3jV69T91k.png" width="1000" height="500">
 </div>
 
+Yi model family is a series of language and multimodal models. It is based on 6B and 34B pretrained language models, then extended to **chat** models, 200K **long context** models, depth-upscaled models (9B), and **vision**-language models.
 
-# Benchmarks
+# ⭐️ Highlights
+
+- **Strong performance**: Yi-1.5-34B is on par with or surpasses GPT-3.5 in commonsense reasoning, college exams, math, coding, reading comprehension, and human preference win-rate on multiple evaluation benchmarks.
+
+- **Cost-effective**: For 6B, 9B, and 34B, you can perform inference on consumer-grade hardware (like the RTX 4090). Additionally, 34B is large enough with complex reasoning and emergent abilities, giving a nice performance-cost balance.
+
+# 📊 Benchmarks
 
 Add Yi-1.5 benchmark results, like
 
-![image/png](https://cdn-uploads.huggingface.co/production/uploads/656d9adce8bf55919aca7c3f/uhGUJd5dNaWL6VnPpctuJ.png)
+xx base models achieve strong performance on a wide range of benchmarks like MMLU, and xx finetuned chat models deliver strong human preference rate on major evaluation platforms like AlpacaEval and Chatbot Arena.
 
+![image/png](https://cdn-uploads.huggingface.co/production/uploads/656d9adce8bf55919aca7c3f/uhGUJd5dNaWL6VnPpctuJ.png)
 
 ![image/png](https://cdn-uploads.huggingface.co/production/uploads/656d9adce8bf55919aca7c3f/kVHWz7yEY3UJlcRD2nwf2.png)
 
-# News
+# 📰 News
+
+- 2024-03-16: The Yi-9B-200K is open-sourced and available to the public.
+
+- 2024-03-08: Yi Tech Report is published!
+
+- 2024-03-07: The long text capability of the Yi-34B-200K has been enhanced.
 
+For complete news history, see [News](xx.md).