diff --git a/litellm/proxy/common_request_processing.py b/litellm/proxy/common_request_processing.py
index 9d69f4c97ba..0b985c6b785 100644
--- a/litellm/proxy/common_request_processing.py
+++ b/litellm/proxy/common_request_processing.py
@@ -923,9 +923,11 @@ class ProxyBaseLLMRequestProcessing:
         ):
             self.data["model"] = user_api_key_dict.aliases[self.data["model"]]
 
-        ### WEB SEARCH REDIRECT (per-deployment force) ###
-        # if request only contains a web_search tool and has any deployment that has
-        # `force_websearch_model`, redirect to that model.
+        ### WEB SEARCH REDIRECT ###
+        # if request only contains web_search tools, check for redirect:
+        # 1. per-deployment force: always redirect to `force_websearch_model`
+        # 2. global fallback: redirect to `websearch_fallback_model` only when
+        #    the current model group does not support web search
         if (
             isinstance(self.data.get("model"), str)
             and llm_router is not None
@@ -939,6 +941,13 @@ class ProxyBaseLLMRequestProcessing:
                 if model := dep.get("litellm_params", {}).get("force_websearch_model"):
                     self.data["model"] = model
                     break
+                if (
+                    (dep_model := dep.get("litellm_params", {}).get("model", ""))
+                    and not litellm.supports_web_search(dep_model)
+                    and (model := getattr(litellm, "websearch_fallback_model", None))
+                ):
+                    self.data["model"] = model
+                    break
 
         self.data["litellm_call_id"] = request.headers.get(
             "x-litellm-call-id", str(uuid.uuid4())
diff --git a/litellm/utils.py b/litellm/utils.py
index 2ba6ef9cae8..9121473918c 100644
--- a/litellm/utils.py
+++ b/litellm/utils.py
@@ -5579,7 +5579,7 @@ def _check_provider_match(model_info: dict, custom_llm_provider: Optional[str])
             # as a last attempt if the model is not on Azure AI, Azure then fallback to OpenAI cost
             # tracking the cost is better than attributing 0 cost to it.
             return True
-        elif custom_llm_provider == "github":
+        elif custom_llm_provider in ("github", "github_copilot"):
             # Allow github/<model> aliases to reuse existing provider metadata.
             return True
         else: