fix(utils.py): predibase exception mapping - map 424 as a badrequest error

litellm/llms/predibase.py

@@ -1,27 +1,26 @@
 # What is this?
 ## Controller file for Predibase Integration - https://predibase.com/
 
-from functools import partial
+import copy
-import os, types
-import traceback
 import json
-from enum import Enum
+import os
-import requests, copy  # type: ignore
 import time
-from typing import Callable, Optional, List, Literal, Union
+import traceback
-from litellm.utils import (
+import types
-    ModelResponse,
+from enum import Enum
-    Usage,
+from functools import partial
-    CustomStreamWrapper,
+from typing import Callable, List, Literal, Optional, Union
-    Message,
+
-    Choices,
-)
-from litellm.litellm_core_utils.core_helpers import map_finish_reason
-import litellm
-from .prompt_templates.factory import prompt_factory, custom_prompt
-from litellm.llms.custom_httpx.http_handler import AsyncHTTPHandler
-from .base import BaseLLM
 import httpx  # type: ignore
+import requests  # type: ignore
+
+import litellm
+from litellm.litellm_core_utils.core_helpers import map_finish_reason
+from litellm.llms.custom_httpx.http_handler import AsyncHTTPHandler
+from litellm.utils import Choices, CustomStreamWrapper, Message, ModelResponse, Usage
+
+from .base import BaseLLM
+from .prompt_templates.factory import custom_prompt, prompt_factory
 
 
 class PredibaseError(Exception):
@@ -496,7 +495,9 @@ class PredibaseChatCompletion(BaseLLM):
         except httpx.HTTPStatusError as e:
             raise PredibaseError(
                 status_code=e.response.status_code,
-                message="HTTPStatusError - {}".format(e.response.text),
+                message="HTTPStatusError - received status_code={}, error_message={}".format(
+                    e.response.status_code, e.response.text
+                ),
             )
         except Exception as e:
             raise PredibaseError(

litellm/proxy/_super_secret_config.yaml

@@ -14,9 +14,10 @@ model_list:
 - model_name: fake-openai-endpoint
   litellm_params:
     model: predibase/llama-3-8b-instruct
-    api_base: "http://0.0.0.0:8000"
+    # api_base: "http://0.0.0.0:8081"
     api_key: os.environ/PREDIBASE_API_KEY
     tenant_id: os.environ/PREDIBASE_TENANT_ID
+    adapter_id: qwoiqjdoqin
     max_retries: 0
     temperature: 0.1
     max_new_tokens: 256
@@ -73,6 +74,8 @@ model_list:
 
 litellm_settings:
   callbacks: ["dynamic_rate_limiter"]
+  # success_callback: ["langfuse"]
+  # failure_callback: ["langfuse"]
   # default_team_settings: 
   #   - team_id: proj1
   #     success_callback: ["langfuse"]

litellm/utils.py

@@ -6157,13 +6157,6 @@ def exception_type(
                         response=original_exception.response,
                         litellm_debug_info=extra_information,
                     )
-                if "Request failed during generation" in error_str:
-                    # this is an internal server error from predibase
-                    raise litellm.InternalServerError(
-                        message=f"PredibaseException - {error_str}",
-                        llm_provider="predibase",
-                        model=model,
-                    )
                 elif hasattr(original_exception, "status_code"):
                     if original_exception.status_code == 500:
                         exception_mapping_worked = True
@@ -6201,7 +6194,10 @@ def exception_type(
                             llm_provider=custom_llm_provider,
                             litellm_debug_info=extra_information,
                         )
-                    elif original_exception.status_code == 422:
+                    elif (
+                        original_exception.status_code == 422
+                        or original_exception.status_code == 424
+                    ):
                         exception_mapping_worked = True
                         raise BadRequestError(
                             message=f"PredibaseException - {original_exception.message}",