(docs) setting soft budgets

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

@@ -79,6 +79,7 @@ curl 'http://0.0.0.0:8000/key/generate' \
   "metadata": {"user": "ishaan@berri.ai"},
   "team_id": "core-infra",
   "max_budget": 10,
+  "soft_budget": 5,
 }'
 ```
 
@@ -93,6 +94,7 @@ Request Params:
 - `config`: *Optional[dict]* - any key-specific configs, overrides config in config.yaml
 - `spend`: *Optional[int]* - Amount spent by key. Default is 0. Will be updated by proxy whenever key is used. https://docs.litellm.ai/docs/proxy/virtual_keys#managing-auth---tracking-spend
 - `max_budget`: *Optional[float]* - Specify max budget for a given key.
+- `soft_budget`: *Optional[float]* - Specify soft limit budget for a given key. Get Alerts when key hits its soft budget
 - `model_max_budget`: *Optional[dict[str, float]]* - Specify max budget for each model, `model_max_budget={"gpt4": 0.5, "gpt-5": 0.01}`
 - `max_parallel_requests`: *Optional[int]* - Rate limit a user based on the number of parallel requests. Raises 429 error, if user's parallel requests > x.
 - `metadata`: *Optional[dict]* - Metadata for key, store information for key. Example metadata = {"team": "core-infra", "app": "app2", "email": "ishaan@berri.ai" }

litellm/__init__.py

@@ -79,6 +79,9 @@ max_budget: float = 0.0  # set the max budget across all providers
 budget_duration: Optional[str] = (
     None  # proxy only - resets budget after fixed duration. You can set duration as seconds ("30s"), minutes ("30m"), hours ("30h"), days ("30d").
 )
+default_soft_budget: float = (
+    50.0  # by default all litellm proxy keys have a soft budget of 50.0
+)
 _openai_finish_reasons = ["stop", "length", "function_call", "content_filter", "null"]
 _openai_completion_params = [
     "functions",

litellm/proxy/proxy_server.py

@@ -1810,6 +1810,9 @@ async def generate_key_helper_fn(
     spend: float,
     key_max_budget: Optional[float] = None,  # key_max_budget is used to Budget Per key
     key_budget_duration: Optional[str] = None,
+    key_soft_budget: Optional[
+        float
+    ] = None,  # key_soft_budget is used to Budget Per key
     max_budget: Optional[float] = None,  # max_budget is used to Budget Per user
     budget_duration: Optional[str] = None,  # max_budget is used to Budget Per user
     token: Optional[str] = None,
@@ -1873,7 +1876,7 @@ async def generate_key_helper_fn(
     if prisma_client is not None:
         # create the Budget Row for the LiteLLM Verification Token
         budget_row = LiteLLM_BudgetTable(
-            soft_budget=50,
+            soft_budget=key_soft_budget or litellm.default_soft_budget,
             model_max_budget=model_max_budget or {},
             created_by=user_id,
             updated_by=user_id,
@@ -3347,6 +3350,8 @@ async def generate_key_fn(
         # if we get max_budget passed to /key/generate, then use it as key_max_budget. Since generate_key_helper_fn is used to make new users
         if "max_budget" in data_json:
             data_json["key_max_budget"] = data_json.pop("max_budget", None)
+        if "soft_budget" in data_json:
+            data_json["key_soft_budget"] = data_json.pop("soft_budget", None)
 
         if "budget_duration" in data_json:
             data_json["key_budget_duration"] = data_json.pop("budget_duration", None)