diff --git a/docs/wip/LOCAL-AI-STACK.md b/docs/wip/LOCAL-AI-STACK.md
index c7b5b62..ce86e1d 100644
--- a/docs/wip/LOCAL-AI-STACK.md
+++ b/docs/wip/LOCAL-AI-STACK.md
@@ -146,17 +146,17 @@ Infrastructure docs are pre-copied to that location.
 
 ---
 
-## Future: Mac Mini M4 Upgrade
+## Future: N5 Air (Ryzen AI 5 255) Upgrade
 
-Planning to migrate AI stack to Mac Mini M4 (32GB):
+Planning to migrate AI stack to N5 Air with Ryzen AI 5 255:
 
-| Metric | N100 (current) | M4 (planned) |
+| Metric | N100 (current) | Ryzen AI 5 255 |
 |--------|----------------|--------------|
-| Speed | ~1 tok/s | ~15-20 tok/s |
-| Max model | 7B | 70B+ |
-| Response time | 30-90s | 3-5s |
+| Speed | ~1 tok/s | ~5-8 tok/s |
+| Max model | 7B | 14B-32B |
+| Response time | 30-90s | 5-15s |
 
-The M4 unified memory architecture is ideal for LLM inference.
+Features XDNA NPU (16 TOPS) for potential AI acceleration. DDR5 + 6c/12t CPU will significantly improve inference.
 
 ---