update config fallbacks

config.yaml

@@ -1,5 +1,5 @@
 model_list:
-  # --- TIER 1: OPENROUTER (3 Akun Load Balancing) ---
+  # --- TIER 1: OPENROUTER ---
   - model_name: primary
     litellm_params:
       model: openrouter/openai/gpt-4o-mini
@@ -13,7 +13,7 @@ model_list:
       model: openrouter/openai/gpt-4o-mini
       api_key: os.environ/OPENROUTER_API_KEY_3
 
-  # --- TIER 2: GROQ (3 Akun Load Balancing) ---
+  # --- TIER 2: GROQ ---
   - model_name: secondary
     litellm_params:
       model: groq/llama-3.3-70b-versatile
@@ -27,7 +27,7 @@ model_list:
       model: groq/llama-3.3-70b-versatile
       api_key: os.environ/GROQ_API_KEY_3
 
-  # --- TIER 3: GITHUB MODELS (3 Akun) ---
+  # --- TIER 3: GITHUB ---
   - model_name: tertiary
     litellm_params:
       model: github/gpt-4o-mini
@@ -43,12 +43,13 @@ model_list:
 
 router_settings:
   routing_strategy: simple-shuffle
-  fallbacks:
-    primary:
-      - secondary
-      - tertiary
   retry_after: 0
   num_retries: 2
   timeout: 60
-  allowed_fails: 3        # per akun — jadi tiap akun boleh fail 3x sebelum cooldown
+  allowed_fails: 3
   cooldown_time: 60
+
+litellm_settings:
+  fallbacks:
+    - {"primary": ["secondary", "tertiary"]}
+    - {"secondary": ["tertiary"]}