Merge pull request #3489 from BerriAI/litellm_send_slack_alerts_model_new

Feat - send slack alerts on `model/new`

litellm/integrations/slack_alerting.py

@@ -412,6 +412,53 @@ class SlackAlerting:
 
         return
 
+    async def model_added_alert(self, model_name: str, litellm_model_name: str):
+        model_info = litellm.model_cost.get(litellm_model_name, {})
+        model_info_str = ""
+        for k, v in model_info.items():
+            if k == "input_cost_per_token" or k == "output_cost_per_token":
+                # when converting to string it should not be 1.63e-06
+                v = "{:.8f}".format(v)
+
+            model_info_str += f"{k}: {v}\n"
+
+        message = f"""
+*🚅 New Model Added*
+Model Name: `{model_name}`
+
+Usage OpenAI Python SDK:
+```
+import openai
+client = openai.OpenAI(
+    api_key="your_api_key",
+    base_url={os.getenv("PROXY_BASE_URL", "http://0.0.0.0:4000")}
+)
+
+response = client.chat.completions.create(
+    model="{model_name}", # model to send to the proxy
+    messages = [
+        {{
+            "role": "user",
+            "content": "this is a test request, write a short poem"
+        }}
+    ]
+)
+```
+
+Model Info: 
+```
+{model_info_str}
+```
+"""
+
+        await self.send_alert(
+            message=message, level="Low", alert_type="new_model_added"
+        )
+        pass
+
+    async def model_removed_alert(self, model_name: str):
+        pass
+
     async def send_alert(
         self,
         message: str,
@@ -422,6 +469,7 @@ class SlackAlerting:
             "llm_requests_hanging",
             "budget_alerts",
             "db_exceptions",
+            "new_model_added",
         ],
     ):
         """
@@ -447,6 +495,9 @@ class SlackAlerting:
         # Get the current timestamp
         current_time = datetime.now().strftime("%H:%M:%S")
         _proxy_base_url = os.getenv("PROXY_BASE_URL", None)
+        if alert_type == "new_model_added":
+            formatted_message = message
+        else:
             formatted_message = (
                 f"Level: `{level}`\nTimestamp: `{current_time}`\n\nMessage: {message}"
             )

litellm/proxy/proxy_server.py

@@ -7303,6 +7303,7 @@ async def add_new_model(
             """
             # encrypt litellm params #
             _litellm_params_dict = model_params.litellm_params.dict(exclude_none=True)
+            _orignal_litellm_model_name = model_params.litellm_params.model
             for k, v in _litellm_params_dict.items():
                 if isinstance(v, str):
                     encrypted_value = encrypt_value(value=v, master_key=master_key)  # type: ignore
@@ -7329,6 +7330,11 @@ async def add_new_model(
                 prisma_client=prisma_client, proxy_logging_obj=proxy_logging_obj
             )
 
+            await proxy_logging_obj.slack_alerting_instance.model_added_alert(
+                model_name=model_params.model_name,
+                litellm_model_name=_orignal_litellm_model_name,
+            )
+
         else:
             raise HTTPException(
                 status_code=500,