diff --git a/strix/config/models.py b/strix/config/models.py
index fd85caef..7f6227f3 100644
--- a/strix/config/models.py
+++ b/strix/config/models.py
@@ -40,12 +40,6 @@ class StrixProvider(MultiProvider):
                 stripped_model_name=stripped_model_name,
             )
         if prefix == "ollama" and stripped_model_name:
-            # Route Ollama through LiteLLM's chat endpoint. The bare ``ollama/``
-            # provider hits ``/api/generate``, which has no function-calling
-            # support, so with ``litellm.drop_params=True`` Strix's tools are
-            # dropped silently and the agent stops after one toolless turn.
-            # ``ollama_chat/`` uses ``/api/chat``, where tool-capable models can
-            # actually drive the agent loop.
             return self._get_fallback_provider("litellm"), f"ollama_chat/{stripped_model_name}"
         return self._get_fallback_provider("litellm"), original_model_name