Update README.md

README.md

@@ -19,20 +19,19 @@ The integration of large language models (LLMs) with function calling has emerge
 ## Main Result
 
 <div align="center">
-  <img src="https://github.com/BingguangHao/FunReason/blob/main/img/result.png?raw=true" width="60%"  />
+  <img src="https://github.com/BingguangHao/FunReason/blob/main/img/result.png?raw=true" width="80%"  />
 </div>
 
 <div align="center">
-  <img src="https://github.com/BingguangHao/FunReason/blob/main/img/code.png?raw=true" width="60%" />
+  <img src="https://github.com/BingguangHao/FunReason/blob/main/img/code.png?raw=true" width="80%" />
 </div>
 
-Performance of SFT and FunReason models on HumanEval and MBPP (including their HumanEval+ and MBPP+) compared with Qwen2.5-Coder-7B-Inst.
 
 ## Usage Recommendations
 
 **We recommend adhering to the following configurations when utilizing the FunReason model, to achieve the expected performance:**
 
-1. **Using the original BFCL system prompt and the chat templete of Qwen.**
+1. **Use the original BFCL system prompt and the chat templete of Qwen.**
 2. In the model handler, the delimiter of the answer is "\n", and the last string obtained by delimiting is taken as the answer
 3. **To ensure that the model engages in thorough reasoning, we recommend enforcing the model to initiate its response with "\<think\>\n" at the beginning of every output.**