fix: restore Python 3.10 imports and response construction

litellm/integrations/websearch_interception/handler.py

@@ -10,9 +10,9 @@ import asyncio
 import math
 import uuid
 from collections.abc import AsyncIterator, Mapping, Sequence
-from typing import TYPE_CHECKING, Any, Final, Literal, Never, TypedDict, TypeVar, cast
+from typing import TYPE_CHECKING, Any, Final, Literal, TypedDict, TypeVar, cast
 
-from typing_extensions import ReadOnly
+from typing_extensions import Never, ReadOnly
 
 import litellm
 from litellm._logging import verbose_logger

litellm/proxy/_experimental/mcp_server/tool_search.py

@@ -3,9 +3,9 @@ from __future__ import annotations
 import json
 from collections.abc import Mapping, Sequence
 from datetime import datetime
-from typing import TYPE_CHECKING, Any, Final, TypedDict, assert_never
+from typing import TYPE_CHECKING, Any, Final, TypedDict
 
-from typing_extensions import ReadOnly, Required
+from typing_extensions import ReadOnly, Required, assert_never
 
 import litellm
 from litellm.proxy.agent_endpoints.agent_search import DEFAULT_AGENT_SEARCH_TOP_K

litellm/proxy/agent_endpoints/endpoints.py

@@ -13,10 +13,10 @@ import os
 import uuid
 from collections.abc import Mapping, Sequence
 from types import MappingProxyType
-from typing import Annotated, Final, TypedDict, assert_never
+from typing import Annotated, Final, TypedDict
 
 from fastapi import APIRouter, Depends, HTTPException, Query, Request
-from typing_extensions import ReadOnly, Required
+from typing_extensions import ReadOnly, Required, assert_never
 
 import litellm
 from litellm._logging import verbose_proxy_logger

litellm/proxy/common_utils/reset_budget_job.py

@@ -7,7 +7,9 @@ from dataclasses import dataclass, field
 from datetime import datetime, timedelta, timezone
 from enum import Enum
 from types import MappingProxyType
-from typing import Final, Literal, Protocol, TypeVar, assert_never
+from typing import Final, Literal, Protocol, TypeVar
+
+from typing_extensions import assert_never
 
 import litellm
 from litellm._logging import verbose_proxy_logger

litellm/proxy/openai_files_endpoints/batch_file_validation.py

@@ -2,7 +2,9 @@ import json
 from collections.abc import Iterator
 from dataclasses import dataclass
 from itertools import chain
-from typing import BinaryIO, Final, NoReturn, assert_never
+from typing import BinaryIO, Final, NoReturn
+
+from typing_extensions import assert_never
 
 from litellm.proxy._types import ProxyException
 

litellm/types/llms/gemini_audio_transcription.py

@@ -1,7 +1,7 @@
-from typing import Literal, Required
+from typing import Literal
 
 from pydantic import BaseModel, ConfigDict
-from typing_extensions import ReadOnly, TypedDict
+from typing_extensions import ReadOnly, Required, TypedDict
 
 
 class GeminiTranscriptionAudioInput(TypedDict):

litellm/types/llms/openai.py

@@ -631,7 +631,7 @@ class ChatCompletionReasoningItem(TypedDict, total=False):
     type: Required[Literal["reasoning"]]
     id: str
     encrypted_content: str | None
-    summary: list["ChatCompletionReasoningSummaryTextBlock"]
+    summary: ReadOnly[list[ChatCompletionReasoningSummaryTextBlock]]
 
 
 class WebSearchOptionsUserLocationApproximate(TypedDict, total=False):

tests/test_litellm/types/llms/test_types_llms_openai.py

@@ -10,6 +10,41 @@ import litellm
 from litellm.types.llms.openai import HttpxBinaryResponseContent
 
 
+@pytest.mark.parametrize("stream", (False, True))
+def test_completion_response_reasoning_summary_round_trip(stream: bool) -> None:
+    from typing import Final
+
+    from litellm.types.llms.openai import (
+        ChatCompletionReasoningItem,
+        ChatCompletionReasoningSummaryTextBlock,
+    )
+    from litellm.types.utils import (
+        Choices,
+        Delta,
+        Message,
+        ModelResponse,
+        ModelResponseStream,
+        StreamingChoices,
+    )
+
+    reasoning_item: Final = ChatCompletionReasoningItem(
+        type="reasoning",
+        id="rs_123",
+        encrypted_content="encrypted",
+        summary=[ChatCompletionReasoningSummaryTextBlock(type="summary_text", text="Reasoning summary")],
+    )
+    response: Final = (
+        ModelResponseStream(choices=[StreamingChoices(delta=Delta(reasoning_items=[reasoning_item]))])
+        if stream
+        else ModelResponse(choices=[Choices(message=Message(reasoning_items=[reasoning_item]))])
+    )
+    message_key: Final = "delta" if stream else "message"
+    assert response.model_dump()["choices"][0][message_key]["reasoning_items"] == [reasoning_item]
+
+    restored: Final = type(response).model_validate_json(response.model_dump_json())
+    assert restored.model_dump()["choices"][0][message_key]["reasoning_items"] == [reasoning_item]
+
+
 def test_generic_event():
     from litellm.types.llms.openai import GenericEvent
 

type-discipline-budget.json

@@ -33,6 +33,6 @@
     "limit": 5535
   },
   "LIT012": {
-    "limit": 4495
+    "limit": 4494
   }
 }