diff --git a/基于AI Agent开发范式的国产GPU大模型推理算子库优化/赛题说明.md b/基于AI Agent开发范式的国产GPU大模型推理算子库优化/赛题说明.md
index 11cf68b..811dcf9 100644
--- a/基于AI Agent开发范式的国产GPU大模型推理算子库优化/赛题说明.md	
+++ b/基于AI Agent开发范式的国产GPU大模型推理算子库优化/赛题说明.md	
@@ -611,7 +611,7 @@ OOM 表示 GPU 显存不足，无法完成当前计算。常见原因：
 
 *   [PyTorch CUDA 编程最佳实践](
 
-教程链接：<a href="https://gitlink.org.cn/metax-maca/op_optimization/tree/master/%E5%9F%BA%E4%BA%8EAI%20Agent%E5%BC%80%E5%8F%91%E8%8C%83%E5%BC%8F%E7%9A%84%E5%9B%BD%E4%BA%A7GPU%E5%A4%A7%E6%A8%A1%E5%9E%8B%E6%8E%A8%E7%90%86%E7%AE%97%E5%AD%90%E5%BA%93%E4%BC%98%E5%8C%96/FlashAttention_Baseline%E5%85%A5%E9%97%A8.md">FlashAttention Baseline 入门</a>
+教程链接：<a href="https://gitlink.org.cn/xiao_xiao/op_optimization/webIDE/tree/master/%E5%9F%BA%E4%BA%8EAI%20Agent%E5%BC%80%E5%8F%91%E8%8C%83%E5%BC%8F%E7%9A%84%E5%9B%BD%E4%BA%A7GPU%E5%A4%A7%E6%A8%A1%E5%9E%8B%E6%8E%A8%E7%90%86%E7%AE%97%E5%AD%90%E5%BA%93%E4%BC%98%E5%8C%96/FlashAttention%E5%85%B3%E9%94%AE%E7%AE%97%E5%AD%90%E8%BF%81%E7%A7%BB%E4%B8%8E%E4%BC%98%E5%8C%96.md">FlashAttention关键算子迁移与优化</a>
 
 ### Track 3：MCTLASS / Fused MoE 算子优化
 
@@ -655,4 +655,4 @@ INT8 量化是一种用于减少模型大小和计算复杂度的方法，特
 
 #### 教程链接：
 
-<a href="https://gitlink.org.cn/metax-maca/op_optimization/tree/master/%E5%9F%BA%E4%BA%8EAI%20Agent%E5%BC%80%E5%8F%91%E8%8C%83%E5%BC%8F%E7%9A%84%E5%9B%BD%E4%BA%A7GPU%E5%A4%A7%E6%A8%A1%E5%9E%8B%E6%8E%A8%E7%90%86%E7%AE%97%E5%AD%90%E5%BA%93%E4%BC%98%E5%8C%96/Fused%20MoE%20Baseline%E5%85%A5%E9%97%A8%EF%BC%9A%E5%BF%AB%E9%80%9F%E8%B7%91%E9%80%9A%E6%9C%80%E5%B0%8F%E9%97%AD%E7%8E%AF%E6%95%99%E7%A8%8B.md">Fused MoE Baseline 入门：快速跑通最小闭环教程</a>
+<a href="https://gitlink.org.cn/xiao_xiao/op_optimization/webIDE/tree/master/%E5%9F%BA%E4%BA%8EAI%20Agent%E5%BC%80%E5%8F%91%E8%8C%83%E5%BC%8F%E7%9A%84%E5%9B%BD%E4%BA%A7GPU%E5%A4%A7%E6%A8%A1%E5%9E%8B%E6%8E%A8%E7%90%86%E7%AE%97%E5%AD%90%E5%BA%93%E4%BC%98%E5%8C%96/MCTLASS_Fused%20MoE%20%E7%AE%97%E5%AD%90%E4%BC%98%E5%8C%96.md">MCTLASS_Fused MoE 算子优化</a>