From 81fd42258cba5c8003541bd2dc95e6d713c5354a Mon Sep 17 00:00:00 2001
From: Krrish Dholakia <krrishdholakia@gmail.com>
Date: Tue, 25 Jun 2024 11:07:07 -0700
Subject: [PATCH] docs(reliability.md): add doc on mock testing fallbacks

---
 docs/my-website/docs/proxy/reliability.md     | 61 +++++++++++++++++++
 ...odel_prices_and_context_window_backup.json | 10 +--
 2 files changed, 66 insertions(+), 5 deletions(-)

diff --git a/docs/my-website/docs/proxy/reliability.md b/docs/my-website/docs/proxy/reliability.md
index a2d24da69b8..c07fc3c26aa 100644
--- a/docs/my-website/docs/proxy/reliability.md
+++ b/docs/my-website/docs/proxy/reliability.md
@@ -431,6 +431,67 @@ litellm_settings:
   content_policy_fallbacks: [{"gpt-3.5-turbo-small": ["claude-opus"]}]
 ```
 
+
+
+### Test Fallbacks! 
+
+Check if your fallbacks are working as expected. 
+
+#### **Regular Fallbacks**
+```bash
+curl -X POST 'http://0.0.0.0:4000/chat/completions' \
+-H 'Content-Type: application/json' \
+-H 'Authorization: Bearer sk-1234' \
+-D '{
+  "model": "my-bad-model",
+  "messages": [
+    {
+      "role": "user",
+      "content": "ping"
+    }
+  ],
+  "mock_testing_fallbacks": true # 👈 KEY CHANGE
+}
+'
+```
+
+#### **Content Policy Fallbacks**
+```bash
+curl -X POST 'http://0.0.0.0:4000/chat/completions' \
+-H 'Content-Type: application/json' \
+-H 'Authorization: Bearer sk-1234' \
+-D '{
+  "model": "my-bad-model",
+  "messages": [
+    {
+      "role": "user",
+      "content": "ping"
+    }
+  ],
+  "mock_testing_content_policy_fallbacks": true # 👈 KEY CHANGE
+}
+'
+```
+
+#### **Context Window Fallbacks**
+
+```bash
+curl -X POST 'http://0.0.0.0:4000/chat/completions' \
+-H 'Content-Type: application/json' \
+-H 'Authorization: Bearer sk-1234' \
+-D '{
+  "model": "my-bad-model",
+  "messages": [
+    {
+      "role": "user",
+      "content": "ping"
+    }
+  ],
+  "mock_testing_context_window_fallbacks": true # 👈 KEY CHANGE
+}
+'
+```
+
 ### EU-Region Filtering (Pre-Call Checks)
 
 **Before call is made** check if a call is within model context window with  **`enable_pre_call_checks: true`**.
diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json
index ef07d87ccb2..415d220f214 100644
--- a/litellm/model_prices_and_context_window_backup.json
+++ b/litellm/model_prices_and_context_window_backup.json
@@ -887,7 +887,7 @@
         "max_input_tokens": 8192,
         "max_output_tokens": 8192,
         "input_cost_per_token": 0.00000005,
-        "output_cost_per_token": 0.00000010,
+        "output_cost_per_token": 0.00000008,
         "litellm_provider": "groq",
         "mode": "chat",
         "supports_function_calling": true
@@ -906,8 +906,8 @@
         "max_tokens": 32768,
         "max_input_tokens": 32768,
         "max_output_tokens": 32768,
-        "input_cost_per_token": 0.00000027,
-        "output_cost_per_token": 0.00000027,
+        "input_cost_per_token": 0.00000024,
+        "output_cost_per_token": 0.00000024,
         "litellm_provider": "groq",
         "mode": "chat",
         "supports_function_calling": true
@@ -916,8 +916,8 @@
         "max_tokens": 8192,
         "max_input_tokens": 8192,
         "max_output_tokens": 8192,
-        "input_cost_per_token": 0.00000010,
-        "output_cost_per_token": 0.00000010,
+        "input_cost_per_token": 0.00000007,
+        "output_cost_per_token": 0.00000007,
         "litellm_provider": "groq",
         "mode": "chat",
         "supports_function_calling": true