Merge branch 'BerriAI:main' into main

docs/my-website/docs/proxy/alerting.md

@@ -1,13 +1,13 @@
-# Slack Alerting
+# 🚨 Alerting 
 
 Get alerts for:
-- hanging LLM api calls
+- Hanging LLM api calls
-- failed LLM api calls
+- Failed LLM api calls
-- slow LLM api calls
+- Slow LLM api calls
-- budget Tracking per key/user:
+- Budget Tracking per key/user:
     - When a User/Key crosses their Budget 
     - When a User/Key is 15% away from crossing their Budget
-- failed db read/writes
+- Failed db read/writes
 
 ## Quick Start
 

docs/my-website/sidebars.js

@@ -43,6 +43,12 @@ const sidebars = {
         "proxy/user_keys",
         "proxy/enterprise",
         "proxy/virtual_keys",
+        "proxy/alerting",
+        {
+          type: "category",
+          label: "Logging",
+          items: ["proxy/logging", "proxy/streaming_logging"],
+        },
         "proxy/team_based_routing",
         "proxy/ui",
         "proxy/cost_tracking",
@@ -58,11 +64,6 @@ const sidebars = {
         "proxy/pii_masking",
         "proxy/prompt_injection",
         "proxy/caching",
-        {
-          type: "category",
-          label: "Logging, Alerting",
-          items: ["proxy/logging", "proxy/alerting", "proxy/streaming_logging"],
-        },
         "proxy/prometheus",
         "proxy/call_hooks",
         "proxy/rules",

litellm/integrations/langfuse.py

@@ -12,7 +12,9 @@ import litellm
 
 class LangFuseLogger:
     # Class variables or attributes
-    def __init__(self, langfuse_public_key=None, langfuse_secret=None):
+    def __init__(
+        self, langfuse_public_key=None, langfuse_secret=None, flush_interval=1
+    ):
         try:
             from langfuse import Langfuse
         except Exception as e:
@@ -31,7 +33,7 @@ class LangFuseLogger:
             host=self.langfuse_host,
             release=self.langfuse_release,
             debug=self.langfuse_debug,
-            flush_interval=1,  # flush interval in seconds
+            flush_interval=flush_interval,  # flush interval in seconds
         )
 
         # set the current langfuse project id in the environ

litellm/integrations/slack_alerting.py

@@ -12,6 +12,7 @@ from litellm.caching import DualCache
 import asyncio
 import aiohttp
 from litellm.llms.custom_httpx.http_handler import AsyncHTTPHandler
+import datetime
 
 
 class SlackAlerting:
@@ -47,6 +48,18 @@ class SlackAlerting:
         self.internal_usage_cache = DualCache()
         self.async_http_handler = AsyncHTTPHandler()
         self.alert_to_webhook_url = alert_to_webhook_url
+        self.langfuse_logger = None
+
+        try:
+            from litellm.integrations.langfuse import LangFuseLogger
+
+            self.langfuse_logger = LangFuseLogger(
+                os.getenv("LANGFUSE_PUBLIC_KEY"),
+                os.getenv("LANGFUSE_SECRET_KEY"),
+                flush_interval=1,
+            )
+        except:
+            pass
 
         pass
 
@@ -93,39 +106,68 @@ class SlackAlerting:
         request_info: str,
         request_data: Optional[dict] = None,
         kwargs: Optional[dict] = None,
+        type: Literal["hanging_request", "slow_response"] = "hanging_request",
+        start_time: Optional[datetime.datetime] = None,
+        end_time: Optional[datetime.datetime] = None,
     ):
         import uuid
 
         # For now: do nothing as we're debugging why this is not working as expected
+        if request_data is not None:
+            trace_id = request_data.get("metadata", {}).get(
+                "trace_id", None
+            )  # get langfuse trace id
+            if trace_id is None:
+                trace_id = "litellm-alert-trace-" + str(uuid.uuid4())
+                request_data["metadata"]["trace_id"] = trace_id
+        elif kwargs is not None:
+            _litellm_params = kwargs.get("litellm_params", {})
+            trace_id = _litellm_params.get("metadata", {}).get(
+                "trace_id", None
+            )  # get langfuse trace id
+            if trace_id is None:
+                trace_id = "litellm-alert-trace-" + str(uuid.uuid4())
+                _litellm_params["metadata"]["trace_id"] = trace_id
+
+        # Log hanging request as an error on langfuse
+        if type == "hanging_request":
+            if self.langfuse_logger is not None:
+                _logging_kwargs = copy.deepcopy(request_data)
+                if _logging_kwargs is None:
+                    _logging_kwargs = {}
+                _logging_kwargs["litellm_params"] = {}
+                request_data = request_data or {}
+                _logging_kwargs["litellm_params"]["metadata"] = request_data.get(
+                    "metadata", {}
+                )
+                # log to langfuse in a separate thread
+                import threading
+
+                threading.Thread(
+                    target=self.langfuse_logger.log_event,
+                    args=(
+                        _logging_kwargs,
+                        None,
+                        start_time,
+                        end_time,
+                        None,
+                        print,
+                        "ERROR",
+                        "Requests is hanging",
+                    ),
+                ).start()
+
+        _langfuse_host = os.environ.get("LANGFUSE_HOST", "https://cloud.langfuse.com")
+        _langfuse_project_id = os.environ.get("LANGFUSE_PROJECT_ID")
+
+        # langfuse urls look like: https://us.cloud.langfuse.com/project/************/traces/litellm-alert-trace-ididi9dk-09292-************
+
+        _langfuse_url = (
+            f"{_langfuse_host}/project/{_langfuse_project_id}/traces/{trace_id}"
+        )
+        request_info += f"\n🪢 Langfuse Trace: {_langfuse_url}"
         return request_info
 
-        # if request_data is not None:
-        #     trace_id = request_data.get("metadata", {}).get(
-        #         "trace_id", None
-        #     )  # get langfuse trace id
-        #     if trace_id is None:
-        #         trace_id = "litellm-alert-trace-" + str(uuid.uuid4())
-        #         request_data["metadata"]["trace_id"] = trace_id
-        # elif kwargs is not None:
-        #     _litellm_params = kwargs.get("litellm_params", {})
-        #     trace_id = _litellm_params.get("metadata", {}).get(
-        #         "trace_id", None
-        #     )  # get langfuse trace id
-        #     if trace_id is None:
-        #         trace_id = "litellm-alert-trace-" + str(uuid.uuid4())
-        #         _litellm_params["metadata"]["trace_id"] = trace_id
-
-        # _langfuse_host = os.environ.get("LANGFUSE_HOST", "https://cloud.langfuse.com")
-        # _langfuse_project_id = os.environ.get("LANGFUSE_PROJECT_ID")
-
-        # # langfuse urls look like: https://us.cloud.langfuse.com/project/************/traces/litellm-alert-trace-ididi9dk-09292-************
-
-        # _langfuse_url = (
-        #     f"{_langfuse_host}/project/{_langfuse_project_id}/traces/{trace_id}"
-        # )
-        # request_info += f"\n🪢 Langfuse Trace: {_langfuse_url}"
-        # return request_info
-
     def _response_taking_too_long_callback(
         self,
         kwargs,  # kwargs to completion
@@ -167,6 +209,14 @@ class SlackAlerting:
             _deployment_latencies = metadata["_latency_per_deployment"]
             if len(_deployment_latencies) == 0:
                 return None
+            try:
+                # try sorting deployments by latency
+                _deployment_latencies = sorted(
+                    _deployment_latencies.items(), key=lambda x: x[1]
+                )
+                _deployment_latencies = dict(_deployment_latencies)
+            except:
+                pass
             for api_base, latency in _deployment_latencies.items():
                 _message_to_send += f"\n{api_base}: {round(latency,2)}s"
             _message_to_send = "```" + _message_to_send + "```"
@@ -194,7 +244,7 @@ class SlackAlerting:
         if time_difference_float > self.alerting_threshold:
             if "langfuse" in litellm.success_callback:
                 request_info = self._add_langfuse_trace_id_to_alert(
-                    request_info=request_info, kwargs=kwargs
+                    request_info=request_info, kwargs=kwargs, type="slow_response"
                 )
             # add deployment latencies to alert
             if (
@@ -222,8 +272,8 @@ class SlackAlerting:
 
     async def response_taking_too_long(
         self,
-        start_time: Optional[float] = None,
+        start_time: Optional[datetime.datetime] = None,
-        end_time: Optional[float] = None,
+        end_time: Optional[datetime.datetime] = None,
         type: Literal["hanging_request", "slow_response"] = "hanging_request",
         request_data: Optional[dict] = None,
     ):
@@ -243,10 +293,6 @@ class SlackAlerting:
             except:
                 messages = ""
             request_info = f"\nRequest Model: `{model}`\nMessages: `{messages}`"
-            if "langfuse" in litellm.success_callback:
-                request_info = self._add_langfuse_trace_id_to_alert(
-                    request_info=request_info, request_data=request_data
-                )
         else:
             request_info = ""
 
@@ -288,6 +334,15 @@ class SlackAlerting:
                     f"`Requests are hanging - {self.alerting_threshold}s+ request time`"
                 )
 
+                if "langfuse" in litellm.success_callback:
+                    request_info = self._add_langfuse_trace_id_to_alert(
+                        request_info=request_info,
+                        request_data=request_data,
+                        type="hanging_request",
+                        start_time=start_time,
+                        end_time=end_time,
+                    )
+
                 # add deployment latencies to alert
                 _deployment_latency_map = self._get_deployment_latencies_to_alert(
                     metadata=request_data.get("metadata", {})

litellm/proxy/_experimental/out/index.html

@@ -1 +1 @@
-<!DOCTYPE html><html id="__next_error__"><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width, initial-scale=1"/><link rel="preload" as="script" fetchPriority="low" href="/ui/_next/static/chunks/webpack-ccae12a25017afa5.js" crossorigin=""/><script src="/ui/_next/static/chunks/fd9d1056-dafd44dfa2da140c.js" async="" crossorigin=""></script><script src="/ui/_next/static/chunks/69-e49705773ae41779.js" async="" crossorigin=""></script><script src="/ui/_next/static/chunks/main-app-9b4fb13a7db53edf.js" async="" crossorigin=""></script><title>LiteLLM Dashboard</title><meta name="description" content="LiteLLM Proxy Admin UI"/><link rel="icon" href="/ui/favicon.ico" type="image/x-icon" sizes="16x16"/><meta name="next-size-adjust"/><script src="/ui/_next/static/chunks/polyfills-c67a75d1b6f99dc8.js" crossorigin="" noModule=""></script></head><body><script src="/ui/_next/static/chunks/webpack-ccae12a25017afa5.js" crossorigin="" async=""></script><script>(self.__next_f=self.__next_f||[]).push([0]);self.__next_f.push([2,null])</script><script>self.__next_f.push([1,"1:HL[\"/ui/_next/static/media/c9a5bc6a7c948fb0-s.p.woff2\",\"font\",{\"crossOrigin\":\"\",\"type\":\"font/woff2\"}]\n2:HL[\"/ui/_next/static/css/5e699db73bf6f8c2.css\",\"style\",{\"crossOrigin\":\"\"}]\n0:\"$L3\"\n"])</script><script>self.__next_f.push([1,"4:I[47690,[],\"\"]\n6:I[77831,[],\"\"]\n7:I[27125,[\"447\",\"static/chunks/447-9f8d32190ff7d16d.js\",\"931\",\"static/chunks/app/page-781ca5f151d78d1d.js\"],\"\"]\n8:I[5613,[],\"\"]\n9:I[31778,[],\"\"]\nb:I[48955,[],\"\"]\nc:[]\n"])</script><script>self.__next_f.push([1,"3:[[[\"$\",\"link\",\"0\",{\"rel\":\"stylesheet\",\"href\":\"/ui/_next/static/css/5e699db73bf6f8c2.css\",\"precedence\":\"next\",\"crossOrigin\":\"\"}]],[\"$\",\"$L4\",null,{\"buildId\":\"PtTtxXIYvdjQsvRgdITlk\",\"assetPrefix\":\"/ui\",\"initialCanonicalUrl\":\"/\",\"initialTree\":[\"\",{\"children\":[\"__PAGE__\",{}]},\"$undefined\",\"$undefined\",true],\"initialSeedData\":[\"\",{\"children\":[\"__PAGE__\",{},[\"$L5\",[\"$\",\"$L6\",null,{\"propsForComponent\":{\"params\":{}},\"Component\":\"$7\",\"isStaticGeneration\":true}],null]]},[null,[\"$\",\"html\",null,{\"lang\":\"en\",\"children\":[\"$\",\"body\",null,{\"className\":\"__className_c23dc8\",\"children\":[\"$\",\"$L8\",null,{\"parallelRouterKey\":\"children\",\"segmentPath\":[\"children\"],\"loading\":\"$undefined\",\"loadingStyles\":\"$undefined\",\"loadingScripts\":\"$undefined\",\"hasLoading\":false,\"error\":\"$undefined\",\"errorStyles\":\"$undefined\",\"errorScripts\":\"$undefined\",\"template\":[\"$\",\"$L9\",null,{}],\"templateStyles\":\"$undefined\",\"templateScripts\":\"$undefined\",\"notFound\":[[\"$\",\"title\",null,{\"children\":\"404: This page could not be found.\"}],[\"$\",\"div\",null,{\"style\":{\"fontFamily\":\"system-ui,\\\"Segoe UI\\\",Roboto,Helvetica,Arial,sans-serif,\\\"Apple Color Emoji\\\",\\\"Segoe UI Emoji\\\"\",\"height\":\"100vh\",\"textAlign\":\"center\",\"display\":\"flex\",\"flexDirection\":\"column\",\"alignItems\":\"center\",\"justifyContent\":\"center\"},\"children\":[\"$\",\"div\",null,{\"children\":[[\"$\",\"style\",null,{\"dangerouslySetInnerHTML\":{\"__html\":\"body{color:#000;background:#fff;margin:0}.next-error-h1{border-right:1px solid rgba(0,0,0,.3)}@media (prefers-color-scheme:dark){body{color:#fff;background:#000}.next-error-h1{border-right:1px solid rgba(255,255,255,.3)}}\"}}],[\"$\",\"h1\",null,{\"className\":\"next-error-h1\",\"style\":{\"display\":\"inline-block\",\"margin\":\"0 20px 0 0\",\"padding\":\"0 23px 0 0\",\"fontSize\":24,\"fontWeight\":500,\"verticalAlign\":\"top\",\"lineHeight\":\"49px\"},\"children\":\"404\"}],[\"$\",\"div\",null,{\"style\":{\"display\":\"inline-block\"},\"children\":[\"$\",\"h2\",null,{\"style\":{\"fontSize\":14,\"fontWeight\":400,\"lineHeight\":\"49px\",\"margin\":0},\"children\":\"This page could not be found.\"}]}]]}]}]],\"notFoundStyles\":[],\"styles\":null}]}]}],null]],\"initialHead\":[false,\"$La\"],\"globalErrorComponent\":\"$b\",\"missingSlots\":\"$Wc\"}]]\n"])</script><script>self.__next_f.push([1,"a:[[\"$\",\"meta\",\"0\",{\"name\":\"viewport\",\"content\":\"width=device-width, initial-scale=1\"}],[\"$\",\"meta\",\"1\",{\"charSet\":\"utf-8\"}],[\"$\",\"title\",\"2\",{\"children\":\"LiteLLM Dashboard\"}],[\"$\",\"meta\",\"3\",{\"name\":\"description\",\"content\":\"LiteLLM Proxy Admin UI\"}],[\"$\",\"link\",\"4\",{\"rel\":\"icon\",\"href\":\"/ui/favicon.ico\",\"type\":\"image/x-icon\",\"sizes\":\"16x16\"}],[\"$\",\"meta\",\"5\",{\"name\":\"next-size-adjust\"}]]\n5:null\n"])</script><script>self.__next_f.push([1,""])</script></body></html>
+<!DOCTYPE html><html id="__next_error__"><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width, initial-scale=1"/><link rel="preload" as="script" fetchPriority="low" href="/ui/_next/static/chunks/webpack-ccae12a25017afa5.js" crossorigin=""/><script src="/ui/_next/static/chunks/fd9d1056-dafd44dfa2da140c.js" async="" crossorigin=""></script><script src="/ui/_next/static/chunks/69-e49705773ae41779.js" async="" crossorigin=""></script><script src="/ui/_next/static/chunks/main-app-9b4fb13a7db53edf.js" async="" crossorigin=""></script><title>LiteLLM Dashboard</title><meta name="description" content="LiteLLM Proxy Admin UI"/><link rel="icon" href="/ui/favicon.ico" type="image/x-icon" sizes="16x16"/><meta name="next-size-adjust"/><script src="/ui/_next/static/chunks/polyfills-c67a75d1b6f99dc8.js" crossorigin="" noModule=""></script></head><body><script src="/ui/_next/static/chunks/webpack-ccae12a25017afa5.js" crossorigin="" async=""></script><script>(self.__next_f=self.__next_f||[]).push([0]);self.__next_f.push([2,null])</script><script>self.__next_f.push([1,"1:HL[\"/ui/_next/static/media/c9a5bc6a7c948fb0-s.p.woff2\",\"font\",{\"crossOrigin\":\"\",\"type\":\"font/woff2\"}]\n2:HL[\"/ui/_next/static/css/5e699db73bf6f8c2.css\",\"style\",{\"crossOrigin\":\"\"}]\n0:\"$L3\"\n"])</script><script>self.__next_f.push([1,"4:I[47690,[],\"\"]\n6:I[77831,[],\"\"]\n7:I[27125,[\"447\",\"static/chunks/447-9f8d32190ff7d16d.js\",\"931\",\"static/chunks/app/page-508c39694bd40fe9.js\"],\"\"]\n8:I[5613,[],\"\"]\n9:I[31778,[],\"\"]\nb:I[48955,[],\"\"]\nc:[]\n"])</script><script>self.__next_f.push([1,"3:[[[\"$\",\"link\",\"0\",{\"rel\":\"stylesheet\",\"href\":\"/ui/_next/static/css/5e699db73bf6f8c2.css\",\"precedence\":\"next\",\"crossOrigin\":\"\"}]],[\"$\",\"$L4\",null,{\"buildId\":\"kbGdRQFfI6W3bEwfzmJDI\",\"assetPrefix\":\"/ui\",\"initialCanonicalUrl\":\"/\",\"initialTree\":[\"\",{\"children\":[\"__PAGE__\",{}]},\"$undefined\",\"$undefined\",true],\"initialSeedData\":[\"\",{\"children\":[\"__PAGE__\",{},[\"$L5\",[\"$\",\"$L6\",null,{\"propsForComponent\":{\"params\":{}},\"Component\":\"$7\",\"isStaticGeneration\":true}],null]]},[null,[\"$\",\"html\",null,{\"lang\":\"en\",\"children\":[\"$\",\"body\",null,{\"className\":\"__className_c23dc8\",\"children\":[\"$\",\"$L8\",null,{\"parallelRouterKey\":\"children\",\"segmentPath\":[\"children\"],\"loading\":\"$undefined\",\"loadingStyles\":\"$undefined\",\"loadingScripts\":\"$undefined\",\"hasLoading\":false,\"error\":\"$undefined\",\"errorStyles\":\"$undefined\",\"errorScripts\":\"$undefined\",\"template\":[\"$\",\"$L9\",null,{}],\"templateStyles\":\"$undefined\",\"templateScripts\":\"$undefined\",\"notFound\":[[\"$\",\"title\",null,{\"children\":\"404: This page could not be found.\"}],[\"$\",\"div\",null,{\"style\":{\"fontFamily\":\"system-ui,\\\"Segoe UI\\\",Roboto,Helvetica,Arial,sans-serif,\\\"Apple Color Emoji\\\",\\\"Segoe UI Emoji\\\"\",\"height\":\"100vh\",\"textAlign\":\"center\",\"display\":\"flex\",\"flexDirection\":\"column\",\"alignItems\":\"center\",\"justifyContent\":\"center\"},\"children\":[\"$\",\"div\",null,{\"children\":[[\"$\",\"style\",null,{\"dangerouslySetInnerHTML\":{\"__html\":\"body{color:#000;background:#fff;margin:0}.next-error-h1{border-right:1px solid rgba(0,0,0,.3)}@media (prefers-color-scheme:dark){body{color:#fff;background:#000}.next-error-h1{border-right:1px solid rgba(255,255,255,.3)}}\"}}],[\"$\",\"h1\",null,{\"className\":\"next-error-h1\",\"style\":{\"display\":\"inline-block\",\"margin\":\"0 20px 0 0\",\"padding\":\"0 23px 0 0\",\"fontSize\":24,\"fontWeight\":500,\"verticalAlign\":\"top\",\"lineHeight\":\"49px\"},\"children\":\"404\"}],[\"$\",\"div\",null,{\"style\":{\"display\":\"inline-block\"},\"children\":[\"$\",\"h2\",null,{\"style\":{\"fontSize\":14,\"fontWeight\":400,\"lineHeight\":\"49px\",\"margin\":0},\"children\":\"This page could not be found.\"}]}]]}]}]],\"notFoundStyles\":[],\"styles\":null}]}]}],null]],\"initialHead\":[false,\"$La\"],\"globalErrorComponent\":\"$b\",\"missingSlots\":\"$Wc\"}]]\n"])</script><script>self.__next_f.push([1,"a:[[\"$\",\"meta\",\"0\",{\"name\":\"viewport\",\"content\":\"width=device-width, initial-scale=1\"}],[\"$\",\"meta\",\"1\",{\"charSet\":\"utf-8\"}],[\"$\",\"title\",\"2\",{\"children\":\"LiteLLM Dashboard\"}],[\"$\",\"meta\",\"3\",{\"name\":\"description\",\"content\":\"LiteLLM Proxy Admin UI\"}],[\"$\",\"link\",\"4\",{\"rel\":\"icon\",\"href\":\"/ui/favicon.ico\",\"type\":\"image/x-icon\",\"sizes\":\"16x16\"}],[\"$\",\"meta\",\"5\",{\"name\":\"next-size-adjust\"}]]\n5:null\n"])</script><script>self.__next_f.push([1,""])</script></body></html>

litellm/proxy/_experimental/out/index.txt

@@ -1,7 +1,7 @@
 2:I[77831,[],""]
-3:I[27125,["447","static/chunks/447-9f8d32190ff7d16d.js","931","static/chunks/app/page-781ca5f151d78d1d.js"],""]
+3:I[27125,["447","static/chunks/447-9f8d32190ff7d16d.js","931","static/chunks/app/page-508c39694bd40fe9.js"],""]
 4:I[5613,[],""]
 5:I[31778,[],""]
-0:["PtTtxXIYvdjQsvRgdITlk",[[["",{"children":["__PAGE__",{}]},"$undefined","$undefined",true],["",{"children":["__PAGE__",{},["$L1",["$","$L2",null,{"propsForComponent":{"params":{}},"Component":"$3","isStaticGeneration":true}],null]]},[null,["$","html",null,{"lang":"en","children":["$","body",null,{"className":"__className_c23dc8","children":["$","$L4",null,{"parallelRouterKey":"children","segmentPath":["children"],"loading":"$undefined","loadingStyles":"$undefined","loadingScripts":"$undefined","hasLoading":false,"error":"$undefined","errorStyles":"$undefined","errorScripts":"$undefined","template":["$","$L5",null,{}],"templateStyles":"$undefined","templateScripts":"$undefined","notFound":[["$","title",null,{"children":"404: This page could not be found."}],["$","div",null,{"style":{"fontFamily":"system-ui,\"Segoe UI\",Roboto,Helvetica,Arial,sans-serif,\"Apple Color Emoji\",\"Segoe UI Emoji\"","height":"100vh","textAlign":"center","display":"flex","flexDirection":"column","alignItems":"center","justifyContent":"center"},"children":["$","div",null,{"children":[["$","style",null,{"dangerouslySetInnerHTML":{"__html":"body{color:#000;background:#fff;margin:0}.next-error-h1{border-right:1px solid rgba(0,0,0,.3)}@media (prefers-color-scheme:dark){body{color:#fff;background:#000}.next-error-h1{border-right:1px solid rgba(255,255,255,.3)}}"}}],["$","h1",null,{"className":"next-error-h1","style":{"display":"inline-block","margin":"0 20px 0 0","padding":"0 23px 0 0","fontSize":24,"fontWeight":500,"verticalAlign":"top","lineHeight":"49px"},"children":"404"}],["$","div",null,{"style":{"display":"inline-block"},"children":["$","h2",null,{"style":{"fontSize":14,"fontWeight":400,"lineHeight":"49px","margin":0},"children":"This page could not be found."}]}]]}]}]],"notFoundStyles":[],"styles":null}]}]}],null]],[[["$","link","0",{"rel":"stylesheet","href":"/ui/_next/static/css/5e699db73bf6f8c2.css","precedence":"next","crossOrigin":""}]],"$L6"]]]]
+0:["kbGdRQFfI6W3bEwfzmJDI",[[["",{"children":["__PAGE__",{}]},"$undefined","$undefined",true],["",{"children":["__PAGE__",{},["$L1",["$","$L2",null,{"propsForComponent":{"params":{}},"Component":"$3","isStaticGeneration":true}],null]]},[null,["$","html",null,{"lang":"en","children":["$","body",null,{"className":"__className_c23dc8","children":["$","$L4",null,{"parallelRouterKey":"children","segmentPath":["children"],"loading":"$undefined","loadingStyles":"$undefined","loadingScripts":"$undefined","hasLoading":false,"error":"$undefined","errorStyles":"$undefined","errorScripts":"$undefined","template":["$","$L5",null,{}],"templateStyles":"$undefined","templateScripts":"$undefined","notFound":[["$","title",null,{"children":"404: This page could not be found."}],["$","div",null,{"style":{"fontFamily":"system-ui,\"Segoe UI\",Roboto,Helvetica,Arial,sans-serif,\"Apple Color Emoji\",\"Segoe UI Emoji\"","height":"100vh","textAlign":"center","display":"flex","flexDirection":"column","alignItems":"center","justifyContent":"center"},"children":["$","div",null,{"children":[["$","style",null,{"dangerouslySetInnerHTML":{"__html":"body{color:#000;background:#fff;margin:0}.next-error-h1{border-right:1px solid rgba(0,0,0,.3)}@media (prefers-color-scheme:dark){body{color:#fff;background:#000}.next-error-h1{border-right:1px solid rgba(255,255,255,.3)}}"}}],["$","h1",null,{"className":"next-error-h1","style":{"display":"inline-block","margin":"0 20px 0 0","padding":"0 23px 0 0","fontSize":24,"fontWeight":500,"verticalAlign":"top","lineHeight":"49px"},"children":"404"}],["$","div",null,{"style":{"display":"inline-block"},"children":["$","h2",null,{"style":{"fontSize":14,"fontWeight":400,"lineHeight":"49px","margin":0},"children":"This page could not be found."}]}]]}]}]],"notFoundStyles":[],"styles":null}]}]}],null]],[[["$","link","0",{"rel":"stylesheet","href":"/ui/_next/static/css/5e699db73bf6f8c2.css","precedence":"next","crossOrigin":""}]],"$L6"]]]]
 6:[["$","meta","0",{"name":"viewport","content":"width=device-width, initial-scale=1"}],["$","meta","1",{"charSet":"utf-8"}],["$","title","2",{"children":"LiteLLM Dashboard"}],["$","meta","3",{"name":"description","content":"LiteLLM Proxy Admin UI"}],["$","link","4",{"rel":"icon","href":"/ui/favicon.ico","type":"image/x-icon","sizes":"16x16"}],["$","meta","5",{"name":"next-size-adjust"}]]
 1:null

litellm/proxy/_super_secret_config.yaml

@@ -4,5 +4,20 @@ model_list:
     api_key: my-fake-key
     model: openai/my-fake-model
   model_name: fake-openai-endpoint
+- litellm_params:
+    api_base: http://0.0.0.0:8080
+    api_key: my-fake-key
+    model: openai/my-fake-model-2
+  model_name: fake-openai-endpoint
+- litellm_params:
+    api_base: http://0.0.0.0:8080
+    api_key: my-fake-key
+    model: openai/my-fake-model-3
+  model_name: fake-openai-endpoint
+- litellm_params:
+    api_base: http://0.0.0.0:8080
+    api_key: my-fake-key
+    model: openai/my-fake-model-4
+  model_name: fake-openai-endpoint
 router_settings:
   num_retries: 0

litellm/proxy/auth/auth_checks.py

@@ -95,7 +95,15 @@ def common_checks(
                 f"'user' param not passed in. 'enforce_user_param'={general_settings['enforce_user_param']}"
             )
     # 7. [OPTIONAL] If 'litellm.max_budget' is set (>0), is proxy under budget
-    if litellm.max_budget > 0 and global_proxy_spend is not None:
+    if (
+        litellm.max_budget > 0
+        and global_proxy_spend is not None
+        # only run global budget checks for OpenAI routes
+        # Reason - the Admin UI should continue working if the proxy crosses it's global budget
+        and route in LiteLLMRoutes.openai_routes.value
+        and route != "/v1/models"
+        and route != "/models"
+    ):
         if global_proxy_spend > litellm.max_budget:
             raise Exception(
                 f"ExceededBudget: LiteLLM Proxy has exceeded its budget. Current spend: {global_proxy_spend}; Max Budget: {litellm.max_budget}"

litellm/router.py

@@ -72,7 +72,9 @@ class Router:
         ## RELIABILITY ##
         num_retries: Optional[int] = None,
         timeout: Optional[float] = None,
-        default_litellm_params={},  # default params for Router.chat.completion.create
+        default_litellm_params: Optional[
+            dict
+        ] = None,  # default params for Router.chat.completion.create
         default_max_parallel_requests: Optional[int] = None,
         set_verbose: bool = False,
         debug_level: Literal["DEBUG", "INFO"] = "INFO",
@@ -158,6 +160,7 @@ class Router:
         router = Router(model_list=model_list, fallbacks=[{"azure-gpt-3.5-turbo": "openai-gpt-3.5-turbo"}])
         ```
         """
+
         if semaphore:
             self.semaphore = semaphore
         self.set_verbose = set_verbose
@@ -260,6 +263,7 @@ class Router:
         )  # dict to store aliases for router, ex. {"gpt-4": "gpt-3.5-turbo"}, all requests with gpt-4 -> get routed to gpt-3.5-turbo group
 
         # make Router.chat.completions.create compatible for openai.chat.completions.create
+        default_litellm_params = default_litellm_params or {}
         self.chat = litellm.Chat(params=default_litellm_params, router_obj=self)
 
         # default litellm args
@@ -475,6 +479,7 @@ class Router:
             )
             kwargs["model_info"] = deployment.get("model_info", {})
             data = deployment["litellm_params"].copy()
+
             model_name = data["model"]
             for k, v in self.default_litellm_params.items():
                 if (
@@ -1453,6 +1458,7 @@ class Router:
                 await asyncio.sleep(timeout)
             elif RouterErrors.user_defined_ratelimit_error.value in str(e):
                 raise e  # don't wait to retry if deployment hits user-defined rate-limit
+
             elif hasattr(original_exception, "status_code") and litellm._should_retry(
                 status_code=original_exception.status_code
             ):
@@ -1614,6 +1620,28 @@ class Router:
                 raise e
             raise original_exception
 
+    def _router_should_retry(
+        self, e: Exception, remaining_retries: int, num_retries: int
+    ):
+        """
+        Calculate back-off, then retry
+        """
+        if hasattr(e, "response") and hasattr(e.response, "headers"):
+            timeout = litellm._calculate_retry_after(
+                remaining_retries=remaining_retries,
+                max_retries=num_retries,
+                response_headers=e.response.headers,
+                min_timeout=self.retry_after,
+            )
+            time.sleep(timeout)
+        else:
+            timeout = litellm._calculate_retry_after(
+                remaining_retries=remaining_retries,
+                max_retries=num_retries,
+                min_timeout=self.retry_after,
+            )
+            time.sleep(timeout)
+
     def function_with_retries(self, *args, **kwargs):
         """
         Try calling the model 3 times. Shuffle between available deployments.
@@ -1633,9 +1661,6 @@ class Router:
             return response
         except Exception as e:
             original_exception = e
-            verbose_router_logger.debug(
-                f"num retries in function with retries: {num_retries}"
-            )
             ### CHECK IF RATE LIMIT / CONTEXT WINDOW ERROR
             if (
                 isinstance(original_exception, litellm.ContextWindowExceededError)
@@ -1649,6 +1674,11 @@ class Router:
             if num_retries > 0:
                 kwargs = self.log_retry(kwargs=kwargs, e=original_exception)
             ### RETRY
+            self._router_should_retry(
+                e=original_exception,
+                remaining_retries=num_retries,
+                num_retries=num_retries,
+            )
             for current_attempt in range(num_retries):
                 verbose_router_logger.debug(
                     f"retrying request. Current attempt - {current_attempt}; retries left: {num_retries}"
@@ -1662,34 +1692,11 @@ class Router:
                     ## LOGGING
                     kwargs = self.log_retry(kwargs=kwargs, e=e)
                     remaining_retries = num_retries - current_attempt
-                    if "No models available" in str(e):
+                    self._router_should_retry(
-                        timeout = litellm._calculate_retry_after(
+                        e=e,
-                            remaining_retries=remaining_retries,
+                        remaining_retries=remaining_retries,
-                            max_retries=num_retries,
+                        num_retries=num_retries,
-                            min_timeout=self.retry_after,
+                    )
-                        )
-                        time.sleep(timeout)
-                    elif (
-                        hasattr(e, "status_code")
-                        and hasattr(e, "response")
-                        and litellm._should_retry(status_code=e.status_code)
-                    ):
-                        if hasattr(e.response, "headers"):
-                            timeout = litellm._calculate_retry_after(
-                                remaining_retries=remaining_retries,
-                                max_retries=num_retries,
-                                response_headers=e.response.headers,
-                                min_timeout=self.retry_after,
-                            )
-                        else:
-                            timeout = litellm._calculate_retry_after(
-                                remaining_retries=remaining_retries,
-                                max_retries=num_retries,
-                                min_timeout=self.retry_after,
-                            )
-                        time.sleep(timeout)
-                    else:
-                        raise e
             raise original_exception
 
     ### HELPER FUNCTIONS
@@ -1987,6 +1994,8 @@ class Router:
                 # check if it ends with a trailing slash
                 if api_base.endswith("/"):
                     api_base += "v1/"
+                elif api_base.endswith("/v1"):
+                    api_base += "/"
                 else:
                     api_base += "/v1/"
 

litellm/tests/test_acooldowns_router.py

@@ -119,7 +119,9 @@ def test_multiple_deployments_parallel():
 
 
 # test_multiple_deployments_parallel()
-def test_cooldown_same_model_name():
+@pytest.mark.parametrize("sync_mode", [True, False])
+@pytest.mark.asyncio
+async def test_cooldown_same_model_name(sync_mode):
     # users could have the same model with different api_base
     # example
     # azure/chatgpt, api_base: 1234
@@ -161,22 +163,40 @@ def test_cooldown_same_model_name():
             num_retries=3,
         )  # type: ignore
 
-        response = router.completion(
+        if sync_mode:
-            model="gpt-3.5-turbo",
+            response = router.completion(
-            messages=[{"role": "user", "content": "hello this request will pass"}],
+                model="gpt-3.5-turbo",
-        )
+                messages=[{"role": "user", "content": "hello this request will pass"}],
-        print(router.model_list)
+            )
-        model_ids = []
+            print(router.model_list)
-        for model in router.model_list:
+            model_ids = []
-            model_ids.append(model["model_info"]["id"])
+            for model in router.model_list:
-        print("\n litellm model ids ", model_ids)
+                model_ids.append(model["model_info"]["id"])
+            print("\n litellm model ids ", model_ids)
 
-        # example litellm_model_names ['azure/chatgpt-v-2-ModelID-64321', 'azure/chatgpt-v-2-ModelID-63960']
+            # example litellm_model_names ['azure/chatgpt-v-2-ModelID-64321', 'azure/chatgpt-v-2-ModelID-63960']
-        assert (
+            assert (
-            model_ids[0] != model_ids[1]
+                model_ids[0] != model_ids[1]
-        )  # ensure both models have a uuid added, and they have different names
+            )  # ensure both models have a uuid added, and they have different names
 
-        print("\ngot response\n", response)
+            print("\ngot response\n", response)
+        else:
+            response = await router.acompletion(
+                model="gpt-3.5-turbo",
+                messages=[{"role": "user", "content": "hello this request will pass"}],
+            )
+            print(router.model_list)
+            model_ids = []
+            for model in router.model_list:
+                model_ids.append(model["model_info"]["id"])
+            print("\n litellm model ids ", model_ids)
+
+            # example litellm_model_names ['azure/chatgpt-v-2-ModelID-64321', 'azure/chatgpt-v-2-ModelID-63960']
+            assert (
+                model_ids[0] != model_ids[1]
+            )  # ensure both models have a uuid added, and they have different names
+
+            print("\ngot response\n", response)
     except Exception as e:
         pytest.fail(f"Got unexpected exception on router! - {e}")
 

litellm/tests/test_embedding.py

@@ -483,6 +483,8 @@ def test_mistral_embeddings():
     except Exception as e:
         pytest.fail(f"Error occurred: {e}")
 
+
+@pytest.mark.skip(reason="local test")
 def test_watsonx_embeddings():
     try:
         litellm.set_verbose = True

litellm/tests/test_least_busy_routing.py

@@ -201,6 +201,7 @@ async def test_router_atext_completion_streaming():
 
 @pytest.mark.asyncio
 async def test_router_completion_streaming():
+    litellm.set_verbose = True
     messages = [
         {"role": "user", "content": "Hello, can you generate a 500 words poem?"}
     ]
@@ -219,9 +220,9 @@ async def test_router_completion_streaming():
         {
             "model_name": "azure-model",
             "litellm_params": {
-                "model": "azure/gpt-35-turbo",
+                "model": "azure/gpt-turbo",
-                "api_key": "os.environ/AZURE_EUROPE_API_KEY",
+                "api_key": "os.environ/AZURE_FRANCE_API_KEY",
-                "api_base": "https://my-endpoint-europe-berri-992.openai.azure.com",
+                "api_base": "https://openai-france-1234.openai.azure.com",
                 "rpm": 6,
             },
             "model_info": {"id": 2},
@@ -229,9 +230,9 @@ async def test_router_completion_streaming():
         {
             "model_name": "azure-model",
             "litellm_params": {
-                "model": "azure/gpt-35-turbo",
+                "model": "azure/gpt-turbo",
-                "api_key": "os.environ/AZURE_CANADA_API_KEY",
+                "api_key": "os.environ/AZURE_FRANCE_API_KEY",
-                "api_base": "https://my-endpoint-canada-berri992.openai.azure.com",
+                "api_base": "https://openai-france-1234.openai.azure.com",
                 "rpm": 6,
             },
             "model_info": {"id": 3},
@@ -262,4 +263,4 @@ async def test_router_completion_streaming():
     ## check if calls equally distributed
     cache_dict = router.cache.get_cache(key=cache_key)
     for k, v in cache_dict.items():
-        assert v == 1
+        assert v == 1, f"Failed. K={k} called v={v} times, cache_dict={cache_dict}"

litellm/tests/test_router.py

@@ -57,6 +57,7 @@ def test_router_num_retries_init(num_retries, max_retries):
     else:
         assert getattr(model_client, "max_retries") == 0
 
+
 @pytest.mark.parametrize(
     "timeout", [10, 1.0, httpx.Timeout(timeout=300.0, connect=20.0)]
 )
@@ -137,6 +138,7 @@ def test_router_azure_ai_studio_init(mistral_api_base):
     print(f"uri_reference: {uri_reference}")
 
     assert "/v1/" in uri_reference
+    assert uri_reference.count("v1") == 1
 
 
 def test_exception_raising():

litellm/tests/test_router_init.py

@@ -203,7 +203,7 @@ def test_timeouts_router():
                 },
             },
         ]
-        router = Router(model_list=model_list)
+        router = Router(model_list=model_list, num_retries=0)
 
         print("PASSED !")
 
@@ -396,7 +396,9 @@ def test_router_init_gpt_4_vision_enhancements():
         pytest.fail(f"Error occurred: {e}")
 
 
-def test_openai_with_organization():
+@pytest.mark.parametrize("sync_mode", [True, False])
+@pytest.mark.asyncio
+async def test_openai_with_organization(sync_mode):
     try:
         print("Testing OpenAI with organization")
         model_list = [
@@ -418,32 +420,65 @@ def test_openai_with_organization():
         print(router.model_list)
         print(router.model_list[0])
 
-        openai_client = router._get_client(
+        if sync_mode:
-            deployment=router.model_list[0],
+            openai_client = router._get_client(
-            kwargs={"input": ["hello"], "model": "openai-bad-org"},
+                deployment=router.model_list[0],
-        )
+                kwargs={"input": ["hello"], "model": "openai-bad-org"},
-        print(vars(openai_client))
-
-        assert openai_client.organization == "org-ikDc4ex8NB"
-
-        # bad org raises error
-
-        try:
-            response = router.completion(
-                model="openai-bad-org",
-                messages=[{"role": "user", "content": "this is a test"}],
             )
-            pytest.fail("Request should have failed - This organization does not exist")
+            print(vars(openai_client))
-        except Exception as e:
-            print("Got exception: " + str(e))
-            assert "No such organization: org-ikDc4ex8NB" in str(e)
 
-        # good org works
+            assert openai_client.organization == "org-ikDc4ex8NB"
-        response = router.completion(
+
-            model="openai-good-org",
+            # bad org raises error
-            messages=[{"role": "user", "content": "this is a test"}],
+
-            max_tokens=5,
+            try:
-        )
+                response = router.completion(
+                    model="openai-bad-org",
+                    messages=[{"role": "user", "content": "this is a test"}],
+                )
+                pytest.fail(
+                    "Request should have failed - This organization does not exist"
+                )
+            except Exception as e:
+                print("Got exception: " + str(e))
+                assert "No such organization: org-ikDc4ex8NB" in str(e)
+
+            # good org works
+            response = router.completion(
+                model="openai-good-org",
+                messages=[{"role": "user", "content": "this is a test"}],
+                max_tokens=5,
+            )
+        else:
+            openai_client = router._get_client(
+                deployment=router.model_list[0],
+                kwargs={"input": ["hello"], "model": "openai-bad-org"},
+                client_type="async",
+            )
+            print(vars(openai_client))
+
+            assert openai_client.organization == "org-ikDc4ex8NB"
+
+            # bad org raises error
+
+            try:
+                response = await router.acompletion(
+                    model="openai-bad-org",
+                    messages=[{"role": "user", "content": "this is a test"}],
+                )
+                pytest.fail(
+                    "Request should have failed - This organization does not exist"
+                )
+            except Exception as e:
+                print("Got exception: " + str(e))
+                assert "No such organization: org-ikDc4ex8NB" in str(e)
+
+            # good org works
+            response = await router.acompletion(
+                model="openai-good-org",
+                messages=[{"role": "user", "content": "this is a test"}],
+                max_tokens=5,
+            )
 
     except Exception as e:
         pytest.fail(f"Error occurred: {e}")

litellm/tests/test_router_timeout.py

@@ -57,6 +57,7 @@ def test_router_timeouts():
         redis_password=os.getenv("REDIS_PASSWORD"),
         redis_port=int(os.getenv("REDIS_PORT")),
         timeout=10,
+        num_retries=0,
     )
 
     print("***** TPM SETTINGS *****")
@@ -89,15 +90,15 @@ def test_router_timeouts():
 
 @pytest.mark.asyncio
 async def test_router_timeouts_bedrock():
-    import openai
+    import openai, uuid
 
     # Model list for OpenAI and Anthropic models
-    model_list = [
+    _model_list = [
         {
             "model_name": "bedrock",
             "litellm_params": {
                 "model": "bedrock/anthropic.claude-instant-v1",
-                "timeout": 0.001,
+                "timeout": 0.00001,
             },
             "tpm": 80000,
         },
@@ -105,17 +106,18 @@ async def test_router_timeouts_bedrock():
 
     # Configure router
     router = Router(
-        model_list=model_list,
+        model_list=_model_list,
         routing_strategy="usage-based-routing",
         debug_level="DEBUG",
         set_verbose=True,
+        num_retries=0,
     )
 
     litellm.set_verbose = True
     try:
         response = await router.acompletion(
             model="bedrock",
-            messages=[{"role": "user", "content": "hello, who are u"}],
+            messages=[{"role": "user", "content": f"hello, who are u {uuid.uuid4()}"}],
         )
         print(response)
         pytest.fail("Did not raise error `openai.APITimeoutError`")

litellm/utils.py

@@ -5487,12 +5487,16 @@ def get_optional_params(
         extra_body = {}
         if "decoding_method" in passed_params:
             extra_body["decoding_method"] = passed_params.pop("decoding_method")
-        if "min_tokens" in passed_params or  "min_new_tokens" in passed_params:
+        if "min_tokens" in passed_params or "min_new_tokens" in passed_params:
-            extra_body["min_new_tokens"] = passed_params.pop("min_tokens", passed_params.pop("min_new_tokens"))
+            extra_body["min_new_tokens"] = passed_params.pop(
+                "min_tokens", passed_params.pop("min_new_tokens")
+            )
         if "top_k" in passed_params:
             extra_body["top_k"] = passed_params.pop("top_k")
         if "truncate_input_tokens" in passed_params:
-            extra_body["truncate_input_tokens"] = passed_params.pop("truncate_input_tokens")
+            extra_body["truncate_input_tokens"] = passed_params.pop(
+                "truncate_input_tokens"
+            )
         if "length_penalty" in passed_params:
             extra_body["length_penalty"] = passed_params.pop("length_penalty")
         if "time_limit" in passed_params:
@@ -5500,7 +5504,7 @@ def get_optional_params(
         if "return_options" in passed_params:
             extra_body["return_options"] = passed_params.pop("return_options")
         optional_params["extra_body"] = (
-            extra_body # openai client supports `extra_body` param
+            extra_body  # openai client supports `extra_body` param
         )
     else:  # assume passing in params for openai/azure openai
         print_verbose(
@@ -9837,19 +9841,21 @@ class CustomStreamWrapper:
             elif isinstance(chunk, (str, bytes)):
                 if isinstance(chunk, bytes):
                     chunk = chunk.decode("utf-8")
-                if 'generated_text' in chunk:
+                if "generated_text" in chunk:
-                    response = chunk.replace('data: ', '').strip()
+                    response = chunk.replace("data: ", "").strip()
                     parsed_response = json.loads(response)
                 else:
                     return {"text": "", "is_finished": False}
             else:
                 print_verbose(f"chunk: {chunk} (Type: {type(chunk)})")
-                raise ValueError(f"Unable to parse response. Original response: {chunk}")
+                raise ValueError(
+                    f"Unable to parse response. Original response: {chunk}"
+                )
             results = parsed_response.get("results", [])
             if len(results) > 0:
                 text = results[0].get("generated_text", "")
                 finish_reason = results[0].get("stop_reason")
-                is_finished = finish_reason != 'not_finished'
+                is_finished = finish_reason != "not_finished"
                 return {
                     "text": text,
                     "is_finished": is_finished,
@@ -10119,16 +10125,6 @@ class CustomStreamWrapper:
             elif self.custom_llm_provider == "watsonx":
                 response_obj = self.handle_watsonx_stream(chunk)
                 completion_obj["content"] = response_obj["text"]
-                print_verbose(f"completion obj content: {completion_obj['content']}")
-                if response_obj.get("prompt_tokens") is not None:
-                    prompt_token_count = getattr(model_response.usage, "prompt_tokens", 0)
-                    model_response.usage.prompt_tokens = (prompt_token_count+response_obj["prompt_tokens"])
-                if response_obj.get("completion_tokens") is not None:
-                    model_response.usage.completion_tokens = response_obj["completion_tokens"]
-                model_response.usage.total_tokens = (
-                    getattr(model_response.usage, "prompt_tokens", 0)
-                    + getattr(model_response.usage, "completion_tokens", 0)
-                )
                 if response_obj["is_finished"]:
                     self.received_finish_reason = response_obj["finish_reason"]
             elif self.custom_llm_provider == "text-completion-openai":

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "litellm"
-version = "1.35.30"
+version = "1.35.31"
 description = "Library to easily interface with LLM API providers"
 authors = ["BerriAI"]
 license = "MIT"
@@ -80,7 +80,7 @@ requires = ["poetry-core", "wheel"]
 build-backend = "poetry.core.masonry.api"
 
 [tool.commitizen]
-version = "1.35.30"
+version = "1.35.31"
 version_files = [
     "pyproject.toml:^version"
 ]

ui/litellm-dashboard/out/index.html

@@ -1 +1 @@
-<!DOCTYPE html><html id="__next_error__"><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width, initial-scale=1"/><link rel="preload" as="script" fetchPriority="low" href="/ui/_next/static/chunks/webpack-ccae12a25017afa5.js" crossorigin=""/><script src="/ui/_next/static/chunks/fd9d1056-dafd44dfa2da140c.js" async="" crossorigin=""></script><script src="/ui/_next/static/chunks/69-e49705773ae41779.js" async="" crossorigin=""></script><script src="/ui/_next/static/chunks/main-app-9b4fb13a7db53edf.js" async="" crossorigin=""></script><title>LiteLLM Dashboard</title><meta name="description" content="LiteLLM Proxy Admin UI"/><link rel="icon" href="/ui/favicon.ico" type="image/x-icon" sizes="16x16"/><meta name="next-size-adjust"/><script src="/ui/_next/static/chunks/polyfills-c67a75d1b6f99dc8.js" crossorigin="" noModule=""></script></head><body><script src="/ui/_next/static/chunks/webpack-ccae12a25017afa5.js" crossorigin="" async=""></script><script>(self.__next_f=self.__next_f||[]).push([0]);self.__next_f.push([2,null])</script><script>self.__next_f.push([1,"1:HL[\"/ui/_next/static/media/c9a5bc6a7c948fb0-s.p.woff2\",\"font\",{\"crossOrigin\":\"\",\"type\":\"font/woff2\"}]\n2:HL[\"/ui/_next/static/css/5e699db73bf6f8c2.css\",\"style\",{\"crossOrigin\":\"\"}]\n0:\"$L3\"\n"])</script><script>self.__next_f.push([1,"4:I[47690,[],\"\"]\n6:I[77831,[],\"\"]\n7:I[27125,[\"447\",\"static/chunks/447-9f8d32190ff7d16d.js\",\"931\",\"static/chunks/app/page-781ca5f151d78d1d.js\"],\"\"]\n8:I[5613,[],\"\"]\n9:I[31778,[],\"\"]\nb:I[48955,[],\"\"]\nc:[]\n"])</script><script>self.__next_f.push([1,"3:[[[\"$\",\"link\",\"0\",{\"rel\":\"stylesheet\",\"href\":\"/ui/_next/static/css/5e699db73bf6f8c2.css\",\"precedence\":\"next\",\"crossOrigin\":\"\"}]],[\"$\",\"$L4\",null,{\"buildId\":\"PtTtxXIYvdjQsvRgdITlk\",\"assetPrefix\":\"/ui\",\"initialCanonicalUrl\":\"/\",\"initialTree\":[\"\",{\"children\":[\"__PAGE__\",{}]},\"$undefined\",\"$undefined\",true],\"initialSeedData\":[\"\",{\"children\":[\"__PAGE__\",{},[\"$L5\",[\"$\",\"$L6\",null,{\"propsForComponent\":{\"params\":{}},\"Component\":\"$7\",\"isStaticGeneration\":true}],null]]},[null,[\"$\",\"html\",null,{\"lang\":\"en\",\"children\":[\"$\",\"body\",null,{\"className\":\"__className_c23dc8\",\"children\":[\"$\",\"$L8\",null,{\"parallelRouterKey\":\"children\",\"segmentPath\":[\"children\"],\"loading\":\"$undefined\",\"loadingStyles\":\"$undefined\",\"loadingScripts\":\"$undefined\",\"hasLoading\":false,\"error\":\"$undefined\",\"errorStyles\":\"$undefined\",\"errorScripts\":\"$undefined\",\"template\":[\"$\",\"$L9\",null,{}],\"templateStyles\":\"$undefined\",\"templateScripts\":\"$undefined\",\"notFound\":[[\"$\",\"title\",null,{\"children\":\"404: This page could not be found.\"}],[\"$\",\"div\",null,{\"style\":{\"fontFamily\":\"system-ui,\\\"Segoe UI\\\",Roboto,Helvetica,Arial,sans-serif,\\\"Apple Color Emoji\\\",\\\"Segoe UI Emoji\\\"\",\"height\":\"100vh\",\"textAlign\":\"center\",\"display\":\"flex\",\"flexDirection\":\"column\",\"alignItems\":\"center\",\"justifyContent\":\"center\"},\"children\":[\"$\",\"div\",null,{\"children\":[[\"$\",\"style\",null,{\"dangerouslySetInnerHTML\":{\"__html\":\"body{color:#000;background:#fff;margin:0}.next-error-h1{border-right:1px solid rgba(0,0,0,.3)}@media (prefers-color-scheme:dark){body{color:#fff;background:#000}.next-error-h1{border-right:1px solid rgba(255,255,255,.3)}}\"}}],[\"$\",\"h1\",null,{\"className\":\"next-error-h1\",\"style\":{\"display\":\"inline-block\",\"margin\":\"0 20px 0 0\",\"padding\":\"0 23px 0 0\",\"fontSize\":24,\"fontWeight\":500,\"verticalAlign\":\"top\",\"lineHeight\":\"49px\"},\"children\":\"404\"}],[\"$\",\"div\",null,{\"style\":{\"display\":\"inline-block\"},\"children\":[\"$\",\"h2\",null,{\"style\":{\"fontSize\":14,\"fontWeight\":400,\"lineHeight\":\"49px\",\"margin\":0},\"children\":\"This page could not be found.\"}]}]]}]}]],\"notFoundStyles\":[],\"styles\":null}]}]}],null]],\"initialHead\":[false,\"$La\"],\"globalErrorComponent\":\"$b\",\"missingSlots\":\"$Wc\"}]]\n"])</script><script>self.__next_f.push([1,"a:[[\"$\",\"meta\",\"0\",{\"name\":\"viewport\",\"content\":\"width=device-width, initial-scale=1\"}],[\"$\",\"meta\",\"1\",{\"charSet\":\"utf-8\"}],[\"$\",\"title\",\"2\",{\"children\":\"LiteLLM Dashboard\"}],[\"$\",\"meta\",\"3\",{\"name\":\"description\",\"content\":\"LiteLLM Proxy Admin UI\"}],[\"$\",\"link\",\"4\",{\"rel\":\"icon\",\"href\":\"/ui/favicon.ico\",\"type\":\"image/x-icon\",\"sizes\":\"16x16\"}],[\"$\",\"meta\",\"5\",{\"name\":\"next-size-adjust\"}]]\n5:null\n"])</script><script>self.__next_f.push([1,""])</script></body></html>
+<!DOCTYPE html><html id="__next_error__"><head><meta charSet="utf-8"/><meta name="viewport" content="width=device-width, initial-scale=1"/><link rel="preload" as="script" fetchPriority="low" href="/ui/_next/static/chunks/webpack-ccae12a25017afa5.js" crossorigin=""/><script src="/ui/_next/static/chunks/fd9d1056-dafd44dfa2da140c.js" async="" crossorigin=""></script><script src="/ui/_next/static/chunks/69-e49705773ae41779.js" async="" crossorigin=""></script><script src="/ui/_next/static/chunks/main-app-9b4fb13a7db53edf.js" async="" crossorigin=""></script><title>LiteLLM Dashboard</title><meta name="description" content="LiteLLM Proxy Admin UI"/><link rel="icon" href="/ui/favicon.ico" type="image/x-icon" sizes="16x16"/><meta name="next-size-adjust"/><script src="/ui/_next/static/chunks/polyfills-c67a75d1b6f99dc8.js" crossorigin="" noModule=""></script></head><body><script src="/ui/_next/static/chunks/webpack-ccae12a25017afa5.js" crossorigin="" async=""></script><script>(self.__next_f=self.__next_f||[]).push([0]);self.__next_f.push([2,null])</script><script>self.__next_f.push([1,"1:HL[\"/ui/_next/static/media/c9a5bc6a7c948fb0-s.p.woff2\",\"font\",{\"crossOrigin\":\"\",\"type\":\"font/woff2\"}]\n2:HL[\"/ui/_next/static/css/5e699db73bf6f8c2.css\",\"style\",{\"crossOrigin\":\"\"}]\n0:\"$L3\"\n"])</script><script>self.__next_f.push([1,"4:I[47690,[],\"\"]\n6:I[77831,[],\"\"]\n7:I[27125,[\"447\",\"static/chunks/447-9f8d32190ff7d16d.js\",\"931\",\"static/chunks/app/page-508c39694bd40fe9.js\"],\"\"]\n8:I[5613,[],\"\"]\n9:I[31778,[],\"\"]\nb:I[48955,[],\"\"]\nc:[]\n"])</script><script>self.__next_f.push([1,"3:[[[\"$\",\"link\",\"0\",{\"rel\":\"stylesheet\",\"href\":\"/ui/_next/static/css/5e699db73bf6f8c2.css\",\"precedence\":\"next\",\"crossOrigin\":\"\"}]],[\"$\",\"$L4\",null,{\"buildId\":\"kbGdRQFfI6W3bEwfzmJDI\",\"assetPrefix\":\"/ui\",\"initialCanonicalUrl\":\"/\",\"initialTree\":[\"\",{\"children\":[\"__PAGE__\",{}]},\"$undefined\",\"$undefined\",true],\"initialSeedData\":[\"\",{\"children\":[\"__PAGE__\",{},[\"$L5\",[\"$\",\"$L6\",null,{\"propsForComponent\":{\"params\":{}},\"Component\":\"$7\",\"isStaticGeneration\":true}],null]]},[null,[\"$\",\"html\",null,{\"lang\":\"en\",\"children\":[\"$\",\"body\",null,{\"className\":\"__className_c23dc8\",\"children\":[\"$\",\"$L8\",null,{\"parallelRouterKey\":\"children\",\"segmentPath\":[\"children\"],\"loading\":\"$undefined\",\"loadingStyles\":\"$undefined\",\"loadingScripts\":\"$undefined\",\"hasLoading\":false,\"error\":\"$undefined\",\"errorStyles\":\"$undefined\",\"errorScripts\":\"$undefined\",\"template\":[\"$\",\"$L9\",null,{}],\"templateStyles\":\"$undefined\",\"templateScripts\":\"$undefined\",\"notFound\":[[\"$\",\"title\",null,{\"children\":\"404: This page could not be found.\"}],[\"$\",\"div\",null,{\"style\":{\"fontFamily\":\"system-ui,\\\"Segoe UI\\\",Roboto,Helvetica,Arial,sans-serif,\\\"Apple Color Emoji\\\",\\\"Segoe UI Emoji\\\"\",\"height\":\"100vh\",\"textAlign\":\"center\",\"display\":\"flex\",\"flexDirection\":\"column\",\"alignItems\":\"center\",\"justifyContent\":\"center\"},\"children\":[\"$\",\"div\",null,{\"children\":[[\"$\",\"style\",null,{\"dangerouslySetInnerHTML\":{\"__html\":\"body{color:#000;background:#fff;margin:0}.next-error-h1{border-right:1px solid rgba(0,0,0,.3)}@media (prefers-color-scheme:dark){body{color:#fff;background:#000}.next-error-h1{border-right:1px solid rgba(255,255,255,.3)}}\"}}],[\"$\",\"h1\",null,{\"className\":\"next-error-h1\",\"style\":{\"display\":\"inline-block\",\"margin\":\"0 20px 0 0\",\"padding\":\"0 23px 0 0\",\"fontSize\":24,\"fontWeight\":500,\"verticalAlign\":\"top\",\"lineHeight\":\"49px\"},\"children\":\"404\"}],[\"$\",\"div\",null,{\"style\":{\"display\":\"inline-block\"},\"children\":[\"$\",\"h2\",null,{\"style\":{\"fontSize\":14,\"fontWeight\":400,\"lineHeight\":\"49px\",\"margin\":0},\"children\":\"This page could not be found.\"}]}]]}]}]],\"notFoundStyles\":[],\"styles\":null}]}]}],null]],\"initialHead\":[false,\"$La\"],\"globalErrorComponent\":\"$b\",\"missingSlots\":\"$Wc\"}]]\n"])</script><script>self.__next_f.push([1,"a:[[\"$\",\"meta\",\"0\",{\"name\":\"viewport\",\"content\":\"width=device-width, initial-scale=1\"}],[\"$\",\"meta\",\"1\",{\"charSet\":\"utf-8\"}],[\"$\",\"title\",\"2\",{\"children\":\"LiteLLM Dashboard\"}],[\"$\",\"meta\",\"3\",{\"name\":\"description\",\"content\":\"LiteLLM Proxy Admin UI\"}],[\"$\",\"link\",\"4\",{\"rel\":\"icon\",\"href\":\"/ui/favicon.ico\",\"type\":\"image/x-icon\",\"sizes\":\"16x16\"}],[\"$\",\"meta\",\"5\",{\"name\":\"next-size-adjust\"}]]\n5:null\n"])</script><script>self.__next_f.push([1,""])</script></body></html>

ui/litellm-dashboard/out/index.txt

@@ -1,7 +1,7 @@
 2:I[77831,[],""]
-3:I[27125,["447","static/chunks/447-9f8d32190ff7d16d.js","931","static/chunks/app/page-781ca5f151d78d1d.js"],""]
+3:I[27125,["447","static/chunks/447-9f8d32190ff7d16d.js","931","static/chunks/app/page-508c39694bd40fe9.js"],""]
 4:I[5613,[],""]
 5:I[31778,[],""]
-0:["PtTtxXIYvdjQsvRgdITlk",[[["",{"children":["__PAGE__",{}]},"$undefined","$undefined",true],["",{"children":["__PAGE__",{},["$L1",["$","$L2",null,{"propsForComponent":{"params":{}},"Component":"$3","isStaticGeneration":true}],null]]},[null,["$","html",null,{"lang":"en","children":["$","body",null,{"className":"__className_c23dc8","children":["$","$L4",null,{"parallelRouterKey":"children","segmentPath":["children"],"loading":"$undefined","loadingStyles":"$undefined","loadingScripts":"$undefined","hasLoading":false,"error":"$undefined","errorStyles":"$undefined","errorScripts":"$undefined","template":["$","$L5",null,{}],"templateStyles":"$undefined","templateScripts":"$undefined","notFound":[["$","title",null,{"children":"404: This page could not be found."}],["$","div",null,{"style":{"fontFamily":"system-ui,\"Segoe UI\",Roboto,Helvetica,Arial,sans-serif,\"Apple Color Emoji\",\"Segoe UI Emoji\"","height":"100vh","textAlign":"center","display":"flex","flexDirection":"column","alignItems":"center","justifyContent":"center"},"children":["$","div",null,{"children":[["$","style",null,{"dangerouslySetInnerHTML":{"__html":"body{color:#000;background:#fff;margin:0}.next-error-h1{border-right:1px solid rgba(0,0,0,.3)}@media (prefers-color-scheme:dark){body{color:#fff;background:#000}.next-error-h1{border-right:1px solid rgba(255,255,255,.3)}}"}}],["$","h1",null,{"className":"next-error-h1","style":{"display":"inline-block","margin":"0 20px 0 0","padding":"0 23px 0 0","fontSize":24,"fontWeight":500,"verticalAlign":"top","lineHeight":"49px"},"children":"404"}],["$","div",null,{"style":{"display":"inline-block"},"children":["$","h2",null,{"style":{"fontSize":14,"fontWeight":400,"lineHeight":"49px","margin":0},"children":"This page could not be found."}]}]]}]}]],"notFoundStyles":[],"styles":null}]}]}],null]],[[["$","link","0",{"rel":"stylesheet","href":"/ui/_next/static/css/5e699db73bf6f8c2.css","precedence":"next","crossOrigin":""}]],"$L6"]]]]
+0:["kbGdRQFfI6W3bEwfzmJDI",[[["",{"children":["__PAGE__",{}]},"$undefined","$undefined",true],["",{"children":["__PAGE__",{},["$L1",["$","$L2",null,{"propsForComponent":{"params":{}},"Component":"$3","isStaticGeneration":true}],null]]},[null,["$","html",null,{"lang":"en","children":["$","body",null,{"className":"__className_c23dc8","children":["$","$L4",null,{"parallelRouterKey":"children","segmentPath":["children"],"loading":"$undefined","loadingStyles":"$undefined","loadingScripts":"$undefined","hasLoading":false,"error":"$undefined","errorStyles":"$undefined","errorScripts":"$undefined","template":["$","$L5",null,{}],"templateStyles":"$undefined","templateScripts":"$undefined","notFound":[["$","title",null,{"children":"404: This page could not be found."}],["$","div",null,{"style":{"fontFamily":"system-ui,\"Segoe UI\",Roboto,Helvetica,Arial,sans-serif,\"Apple Color Emoji\",\"Segoe UI Emoji\"","height":"100vh","textAlign":"center","display":"flex","flexDirection":"column","alignItems":"center","justifyContent":"center"},"children":["$","div",null,{"children":[["$","style",null,{"dangerouslySetInnerHTML":{"__html":"body{color:#000;background:#fff;margin:0}.next-error-h1{border-right:1px solid rgba(0,0,0,.3)}@media (prefers-color-scheme:dark){body{color:#fff;background:#000}.next-error-h1{border-right:1px solid rgba(255,255,255,.3)}}"}}],["$","h1",null,{"className":"next-error-h1","style":{"display":"inline-block","margin":"0 20px 0 0","padding":"0 23px 0 0","fontSize":24,"fontWeight":500,"verticalAlign":"top","lineHeight":"49px"},"children":"404"}],["$","div",null,{"style":{"display":"inline-block"},"children":["$","h2",null,{"style":{"fontSize":14,"fontWeight":400,"lineHeight":"49px","margin":0},"children":"This page could not be found."}]}]]}]}]],"notFoundStyles":[],"styles":null}]}]}],null]],[[["$","link","0",{"rel":"stylesheet","href":"/ui/_next/static/css/5e699db73bf6f8c2.css","precedence":"next","crossOrigin":""}]],"$L6"]]]]
 6:[["$","meta","0",{"name":"viewport","content":"width=device-width, initial-scale=1"}],["$","meta","1",{"charSet":"utf-8"}],["$","title","2",{"children":"LiteLLM Dashboard"}],["$","meta","3",{"name":"description","content":"LiteLLM Proxy Admin UI"}],["$","link","4",{"rel":"icon","href":"/ui/favicon.ico","type":"image/x-icon","sizes":"16x16"}],["$","meta","5",{"name":"next-size-adjust"}]]
 1:null

ui/litellm-dashboard/src/components/create_key_button.tsx

@@ -39,6 +39,7 @@ const CreateKey: React.FC<CreateKeyProps> = ({
   const [apiKey, setApiKey] = useState(null);
   const [softBudget, setSoftBudget] = useState(null);
   const [userModels, setUserModels] = useState([]);
+  const [modelsToPick, setModelsToPick] = useState([]);
   const handleOk = () => {
     setIsModalVisible(false);
     form.resetFields();
@@ -95,6 +96,30 @@ const CreateKey: React.FC<CreateKeyProps> = ({
     message.success('API Key copied to clipboard');
 };
 
+  useEffect(() => {
+    let tempModelsToPick = [];
+
+    if (team) {
+      if (team.models.length > 0) {
+        if (team.models.includes("all-proxy-models")) {
+          // if the team has all-proxy-models show all available models
+          tempModelsToPick = userModels;
+        } else {
+          // show team models
+          tempModelsToPick = team.models;
+        }
+      } else {
+        // show all available models if the team has no models set
+        tempModelsToPick = userModels;
+      }
+    } else {
+      // no team set, show all available models
+      tempModelsToPick = userModels;
+    }
+
+    setModelsToPick(tempModelsToPick);
+  }, [team, userModels]);
+  
 
   return (
     <div>
@@ -161,30 +186,15 @@ const CreateKey: React.FC<CreateKeyProps> = ({
                     <Option key="all-team-models" value="all-team-models">
                       All Team Models
                     </Option>
-                    {team && team.models ? (
+                    {
-                      team.models.includes("all-proxy-models") ? (
+                      modelsToPick.map((model: string) => (
-                        userModels.map((model: string) => (
                           (
                             <Option key={model} value={model}>
                               {model}
                             </Option>
                           )
                         ))
-                      ) : (
+                    }
-                        team.models.map((model: string) => (
-                          <Option key={model} value={model}>
-                            {model}
-                          </Option>
-                        ))
-                      )
-                    ) : (
-                      userModels.map((model: string) => (
-                        <Option key={model} value={model}>
-                          {model}
-                        </Option>
-                      ))
-                    )}
-
                 </Select>
               </Form.Item>
               <Accordion className="mt-20 mb-8" >