delete image

Author: yinmin

website/blog/2025-10-30-milvus.md

@@ -258,15 +258,14 @@ sys     0m0.021s
 
 This test demonstrates Semantic Router's semantic caching in action. By leveraging Milvus as the vector database, it efficiently matches semantically similar queries, improving response times when users ask the same or similar questions.
 
-![performance-comparison](/img/performance-comparison.png)
+
 
 ## Conclusion
 
 As AI workloads grow and cost optimization becomes essential, the combination of vLLM Semantic Router and Milvus provides a practical way to scale intelligently. By routing each query to the right model and caching semantically similar results with a distributed vector database, this setup cuts compute overhead while keeping responses fast and consistent across use cases.
 
 In short, you get smarter scaling—less brute force, more brains.
 
-![smart-scaling](/img/smart-scaling.png)
 
 ---