Update README.md

README.md

@@ -32,6 +32,7 @@ It is designed for **fast CPU inference** with [ONNX Runtime](https://onnxruntim
 | **Supported HW**    | CPU (optimized for Intel AVX512-VNNI, fallback to AVX2)               |
 | **License**         | Apache-2.0                                                            |
 
+---
 
 ## 🚀 Features
 
@@ -40,6 +41,8 @@ It is designed for **fast CPU inference** with [ONNX Runtime](https://onnxruntim
 - 🔄 **Drop-in replacement** — embeddings compatible with the FP32 version.  
 - 🌍 **Multilingual** — supports Russian 🇷🇺 and English 🇬🇧.  
 
+---
+
 ## 🧠 Intended Use
 
 **✅ Recommended for:**
@@ -50,7 +53,9 @@ It is designed for **fast CPU inference** with [ONNX Runtime](https://onnxruntim
 
 **❌ Not ideal for:**
 - Absolute maximum accuracy scenarios (INT8 introduces minor loss)  
-- GPU-optimized pipelines (prefer FP16/FP32 models instead)  
+- GPU-optimized pipelines (prefer FP16/FP32 models instead)
+
+---
 
 ## ⚖️ Pros & Cons of Quantized ONNX
 
@@ -64,6 +69,8 @@ It is designed for **fast CPU inference** with [ONNX Runtime](https://onnxruntim
 - AVX512 optimizations only on modern Intel CPUs.  
 - No GPU acceleration in this export.
 
+---
+
 ## 📊 Benchmark
 
 | Metric                       | Value         |
@@ -75,6 +82,7 @@ It is designed for **fast CPU inference** with [ONNX Runtime](https://onnxruntim
 | Inference speed                 | ~2× faster |
 | Model size (MB)                 | 347.5      |
 
+---
 
 ## 📂 Files
 
@@ -84,6 +92,7 @@ tokenizer.json, vocab.txt, special_tokens_map.json — tokenizer
 
 config.json — model config
 
+---
 
 ## 🧩 Examples