diff --git a/backend/routers/gateway_proxy.py b/backend/routers/gateway_proxy.py
index c8c83fe..8d12a17 100644
--- a/backend/routers/gateway_proxy.py
+++ b/backend/routers/gateway_proxy.py
@@ -41,10 +41,10 @@ _LANE_LABELS = {"coding": "Coding (Router → coder/heavy/fast)", "chat": "Chat
 
 
 @router.get("/models")
-async def models():
-    async with httpx.AsyncClient(timeout=10) as c:
-        r = await c.get(f"{LLAMA_SWAP_URL}/v1/models")
-        data = r.json()
+async def models(request: Request):
+    client = request.app.state.gw_client  # geteilter Keep-Alive-Client (siehe app.py lifespan)
+    r = await client.get(f"{LLAMA_SWAP_URL}/v1/models", timeout=10.0)
+    data = r.json()
     # Lanes ganz oben einblenden, damit IDEs einfach „coding"/„chat" wählen können.
     lanes = [{"id": lane, "object": "model", "owned_by": "mc2-router",
               "description": _LANE_LABELS.get(lane, lane)} for lane in LANES]