Update blog

Author: cmdr2

content/blog/2025-11-07-1762514507.md

@@ -15,7 +15,11 @@ Wrote a simple script to convert ONNX to GGML. It auto-generates C++ code that c
 
 The generated file can work on multiple backends: CPU, CUDA, ROCm, Vulkan, Metal etc, by providing the correct compiler flags during `cmake -B`, e.g. `-D GGML_CUDA=1` for CUDA.
 
-The original PyTorch model was:
+Repo: [https://github.com/cmdr2/graph-compiler](https://github.com/cmdr2/graph-compiler)
+
+I've currently got it to work for a TinyCNN model, and will add more operators as I make it convert larger models.
+
+The PyTorch model used was:
 
 ```py
 class TinyCNN(nn.Module):
@@ -30,8 +34,4 @@ class TinyCNN(nn.Module):
         x = self.relu(self.conv(x))
         x = self.pool(x).flatten(1)
         return self.fc(x)
-```
-
-Repo: [https://github.com/cmdr2/graph-compiler](https://github.com/cmdr2/graph-compiler)
-
-I've currently got it to work for a TinyCNN model, and will add more operators as I make it convert larger models.
+```