diff --git a/litellm/llms/prompt_templates/factory.py b/litellm/llms/prompt_templates/factory.py
index a6d1d64386f..1a576f43a33 100644
--- a/litellm/llms/prompt_templates/factory.py
+++ b/litellm/llms/prompt_templates/factory.py
@@ -1359,11 +1359,8 @@ def prompt_factory(
             "meta-llama/llama-3" in model or "meta-llama-3" in model
         ) and "instruct" in model:
             return hf_chat_template(
-                model=model,
+                model="meta-llama/Meta-Llama-3-8B-Instruct",
                 messages=messages,
-                chat_template=known_tokenizer_config[  # type: ignore
-                    "meta-llama/Meta-Llama-3-8B-Instruct"
-                ]["tokenizer"]["chat_template"],
             )
         elif (
             "tiiuae/falcon" in model
diff --git a/litellm/tests/test_router_fallbacks.py b/litellm/tests/test_router_fallbacks.py
index 98a2449f06b..51d9451a87e 100644
--- a/litellm/tests/test_router_fallbacks.py
+++ b/litellm/tests/test_router_fallbacks.py
@@ -258,6 +258,7 @@ def test_sync_fallbacks_embeddings():
         model_list=model_list,
         fallbacks=[{"bad-azure-embedding-model": ["good-azure-embedding-model"]}],
         set_verbose=False,
+        num_retries=0,
     )
     customHandler = MyCustomHandler()
     litellm.callbacks = [customHandler]
@@ -393,7 +394,7 @@ def test_dynamic_fallbacks_sync():
             },
         ]
 
-        router = Router(model_list=model_list, set_verbose=True)
+        router = Router(model_list=model_list, set_verbose=True, num_retries=0)
         kwargs = {}
         kwargs["model"] = "azure/gpt-3.5-turbo"
         kwargs["messages"] = [{"role": "user", "content": "Hey, how's it going?"}]