diff --git a/.release-please-manifest.json b/.release-please-manifest.json
index 3b81c9b87e..102fa47016 100644
--- a/.release-please-manifest.json
+++ b/.release-please-manifest.json
@@ -1,3 +1,3 @@
 {
-  ".": "1.107.3"
+  ".": "1.108.0"
 }
\ No newline at end of file
diff --git a/.stats.yml b/.stats.yml
index 905a02c44a..2dd0aef46a 100644
--- a/.stats.yml
+++ b/.stats.yml
@@ -1,4 +1,4 @@
 configured_endpoints: 118
-openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai%2Fopenai-d30ff992a48873c1466c49f3c01f2ec8933faebff23424748f8d056065b1bcef.yml
-openapi_spec_hash: e933ec43b46f45c348adb78840e5808d
-config_hash: bf45940f0a7805b4ec2017eecdd36893
+openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/openai%2Fopenai-380330a93b5d010391ca3b36ea193c5353b0dfdf2ddd02789ef84a84ce427e82.yml
+openapi_spec_hash: 859703234259ecdd2a3c6f4de88eb504
+config_hash: b619b45c1e7facf819f902dee8fa4f97
diff --git a/CHANGELOG.md b/CHANGELOG.md
index b9314bd48a..1e35189611 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,5 +1,18 @@
 # Changelog
 
+## 1.108.0 (2025-09-17)
+
+Full Changelog: [v1.107.3...v1.108.0](https://github.com/openai/openai-python/compare/v1.107.3...v1.108.0)
+
+### Features
+
+* **api:** type updates for conversations, reasoning_effort and results for evals ([c2ee28c](https://github.com/openai/openai-python/commit/c2ee28c1b77eed98766fbb01cf1ad2ee240f412e))
+
+
+### Chores
+
+* **internal:** update pydantic dependency ([369d10a](https://github.com/openai/openai-python/commit/369d10a40dfe744f6bfc10c99eb1f58176500120))
+
 ## 1.107.3 (2025-09-15)
 
 Full Changelog: [v1.107.2...v1.107.3](https://github.com/openai/openai-python/compare/v1.107.2...v1.107.3)
diff --git a/api.md b/api.md
index 73b8427387..6bbb47f78c 100644
--- a/api.md
+++ b/api.md
@@ -991,22 +991,17 @@ Types:
 ```python
 from openai.types.conversations import (
     ComputerScreenshotContent,
-    ContainerFileCitationBody,
     Conversation,
     ConversationDeleted,
     ConversationDeletedResource,
-    FileCitationBody,
-    InputFileContent,
-    InputImageContent,
-    InputTextContent,
-    LobProb,
     Message,
-    OutputTextContent,
-    RefusalContent,
     SummaryTextContent,
     TextContent,
-    TopLogProb,
-    URLCitationBody,
+    InputTextContent,
+    OutputTextContent,
+    RefusalContent,
+    InputImageContent,
+    InputFileContent,
 )
 ```
 
diff --git a/pyproject.toml b/pyproject.toml
index 190542e6dc..058b7cda6c 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -1,6 +1,6 @@
 [project]
 name = "openai"
-version = "1.107.3"
+version = "1.108.0"
 description = "The official Python library for the openai API"
 dynamic = ["readme"]
 license = "Apache-2.0"
diff --git a/requirements-dev.lock b/requirements-dev.lock
index eaf136f7e6..0bd1c2c70f 100644
--- a/requirements-dev.lock
+++ b/requirements-dev.lock
@@ -108,6 +108,7 @@ multidict==6.5.0
 mypy==1.14.1
 mypy-extensions==1.0.0
     # via mypy
+nest-asyncio==1.6.0
 nodeenv==1.8.0
     # via pyright
 nox==2023.4.22
@@ -133,11 +134,11 @@ portalocker==2.10.1
 propcache==0.3.2
     # via aiohttp
     # via yarl
-pycparser==2.22
+pycparser==2.23
     # via cffi
-pydantic==2.10.3
+pydantic==2.11.9
     # via openai
-pydantic-core==2.27.1
+pydantic-core==2.33.2
     # via pydantic
 pygments==2.18.0
     # via pytest
@@ -199,6 +200,9 @@ typing-extensions==4.12.2
     # via pydantic
     # via pydantic-core
     # via pyright
+    # via typing-inspection
+typing-inspection==0.4.1
+    # via pydantic
 tzdata==2024.1
     # via pandas
 urllib3==2.2.1
diff --git a/requirements.lock b/requirements.lock
index 3b6ece87e2..a2b6845942 100644
--- a/requirements.lock
+++ b/requirements.lock
@@ -67,11 +67,11 @@ pandas-stubs==2.2.2.240807
 propcache==0.3.2
     # via aiohttp
     # via yarl
-pycparser==2.22
+pycparser==2.23
     # via cffi
-pydantic==2.10.3
+pydantic==2.11.9
     # via openai
-pydantic-core==2.27.1
+pydantic-core==2.33.2
     # via pydantic
 python-dateutil==2.9.0.post0
     # via pandas
@@ -93,7 +93,10 @@ typing-extensions==4.12.2
     # via openai
     # via pydantic
     # via pydantic-core
-tzdata==2024.1
+    # via typing-inspection
+typing-inspection==0.4.1
+    # via pydantic
+tzdata==2025.2
     # via pandas
 websockets==15.0.1
     # via openai
diff --git a/src/openai/_models.py b/src/openai/_models.py
index 8ee8612d1e..af71a91850 100644
--- a/src/openai/_models.py
+++ b/src/openai/_models.py
@@ -281,7 +281,7 @@ def model_dump(
             mode: Literal["json", "python"] | str = "python",
             include: IncEx | None = None,
             exclude: IncEx | None = None,
-            by_alias: bool = False,
+            by_alias: bool | None = None,
             exclude_unset: bool = False,
             exclude_defaults: bool = False,
             exclude_none: bool = False,
@@ -289,6 +289,7 @@ def model_dump(
             warnings: bool | Literal["none", "warn", "error"] = True,
             context: dict[str, Any] | None = None,
             serialize_as_any: bool = False,
+            fallback: Callable[[Any], Any] | None = None,
         ) -> dict[str, Any]:
             """Usage docs: https://docs.pydantic.dev/2.4/concepts/serialization/#modelmodel_dump
 
@@ -320,10 +321,12 @@ def model_dump(
                 raise ValueError("context is only supported in Pydantic v2")
             if serialize_as_any != False:
                 raise ValueError("serialize_as_any is only supported in Pydantic v2")
+            if fallback is not None:
+                raise ValueError("fallback is only supported in Pydantic v2")
             dumped = super().dict(  # pyright: ignore[reportDeprecated]
                 include=include,
                 exclude=exclude,
-                by_alias=by_alias,
+                by_alias=by_alias if by_alias is not None else False,
                 exclude_unset=exclude_unset,
                 exclude_defaults=exclude_defaults,
                 exclude_none=exclude_none,
@@ -338,13 +341,14 @@ def model_dump_json(
             indent: int | None = None,
             include: IncEx | None = None,
             exclude: IncEx | None = None,
-            by_alias: bool = False,
+            by_alias: bool | None = None,
             exclude_unset: bool = False,
             exclude_defaults: bool = False,
             exclude_none: bool = False,
             round_trip: bool = False,
             warnings: bool | Literal["none", "warn", "error"] = True,
             context: dict[str, Any] | None = None,
+            fallback: Callable[[Any], Any] | None = None,
             serialize_as_any: bool = False,
         ) -> str:
             """Usage docs: https://docs.pydantic.dev/2.4/concepts/serialization/#modelmodel_dump_json
@@ -373,11 +377,13 @@ def model_dump_json(
                 raise ValueError("context is only supported in Pydantic v2")
             if serialize_as_any != False:
                 raise ValueError("serialize_as_any is only supported in Pydantic v2")
+            if fallback is not None:
+                raise ValueError("fallback is only supported in Pydantic v2")
             return super().json(  # type: ignore[reportDeprecated]
                 indent=indent,
                 include=include,
                 exclude=exclude,
-                by_alias=by_alias,
+                by_alias=by_alias if by_alias is not None else False,
                 exclude_unset=exclude_unset,
                 exclude_defaults=exclude_defaults,
                 exclude_none=exclude_none,
diff --git a/src/openai/_version.py b/src/openai/_version.py
index aa7660d137..7030fe068c 100644
--- a/src/openai/_version.py
+++ b/src/openai/_version.py
@@ -1,4 +1,4 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
 __title__ = "openai"
-__version__ = "1.107.3"  # x-release-please-version
+__version__ = "1.108.0"  # x-release-please-version
diff --git a/src/openai/types/conversations/__init__.py b/src/openai/types/conversations/__init__.py
index 538966db4f..9dec848737 100644
--- a/src/openai/types/conversations/__init__.py
+++ b/src/openai/types/conversations/__init__.py
@@ -3,15 +3,11 @@
 from __future__ import annotations
 
 from .message import Message as Message
-from .lob_prob import LobProb as LobProb
 from .conversation import Conversation as Conversation
 from .text_content import TextContent as TextContent
-from .top_log_prob import TopLogProb as TopLogProb
 from .refusal_content import RefusalContent as RefusalContent
 from .item_list_params import ItemListParams as ItemListParams
 from .conversation_item import ConversationItem as ConversationItem
-from .url_citation_body import URLCitationBody as URLCitationBody
-from .file_citation_body import FileCitationBody as FileCitationBody
 from .input_file_content import InputFileContent as InputFileContent
 from .input_text_content import InputTextContent as InputTextContent
 from .item_create_params import ItemCreateParams as ItemCreateParams
@@ -19,9 +15,13 @@
 from .output_text_content import OutputTextContent as OutputTextContent
 from .item_retrieve_params import ItemRetrieveParams as ItemRetrieveParams
 from .summary_text_content import SummaryTextContent as SummaryTextContent
+from .refusal_content_param import RefusalContentParam as RefusalContentParam
 from .conversation_item_list import ConversationItemList as ConversationItemList
+from .input_file_content_param import InputFileContentParam as InputFileContentParam
+from .input_text_content_param import InputTextContentParam as InputTextContentParam
+from .input_image_content_param import InputImageContentParam as InputImageContentParam
+from .output_text_content_param import OutputTextContentParam as OutputTextContentParam
 from .conversation_create_params import ConversationCreateParams as ConversationCreateParams
 from .conversation_update_params import ConversationUpdateParams as ConversationUpdateParams
 from .computer_screenshot_content import ComputerScreenshotContent as ComputerScreenshotContent
-from .container_file_citation_body import ContainerFileCitationBody as ContainerFileCitationBody
 from .conversation_deleted_resource import ConversationDeletedResource as ConversationDeletedResource
diff --git a/src/openai/types/conversations/container_file_citation_body.py b/src/openai/types/conversations/container_file_citation_body.py
deleted file mode 100644
index ea460df2e2..0000000000
--- a/src/openai/types/conversations/container_file_citation_body.py
+++ /dev/null
@@ -1,27 +0,0 @@
-# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
-
-from typing_extensions import Literal
-
-from ..._models import BaseModel
-
-__all__ = ["ContainerFileCitationBody"]
-
-
-class ContainerFileCitationBody(BaseModel):
-    container_id: str
-    """The ID of the container file."""
-
-    end_index: int
-    """The index of the last character of the container file citation in the message."""
-
-    file_id: str
-    """The ID of the file."""
-
-    filename: str
-    """The filename of the container file cited."""
-
-    start_index: int
-    """The index of the first character of the container file citation in the message."""
-
-    type: Literal["container_file_citation"]
-    """The type of the container file citation. Always `container_file_citation`."""
diff --git a/src/openai/types/conversations/file_citation_body.py b/src/openai/types/conversations/file_citation_body.py
deleted file mode 100644
index ea90ae381d..0000000000
--- a/src/openai/types/conversations/file_citation_body.py
+++ /dev/null
@@ -1,21 +0,0 @@
-# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
-
-from typing_extensions import Literal
-
-from ..._models import BaseModel
-
-__all__ = ["FileCitationBody"]
-
-
-class FileCitationBody(BaseModel):
-    file_id: str
-    """The ID of the file."""
-
-    filename: str
-    """The filename of the file cited."""
-
-    index: int
-    """The index of the file in the list of files."""
-
-    type: Literal["file_citation"]
-    """The type of the file citation. Always `file_citation`."""
diff --git a/src/openai/types/conversations/input_file_content.py b/src/openai/types/conversations/input_file_content.py
index 6aef7a89d9..ca555d85fc 100644
--- a/src/openai/types/conversations/input_file_content.py
+++ b/src/openai/types/conversations/input_file_content.py
@@ -1,22 +1,7 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
-from typing import Optional
-from typing_extensions import Literal
-
-from ..._models import BaseModel
+from ..responses.response_input_file import ResponseInputFile
 
 __all__ = ["InputFileContent"]
 
-
-class InputFileContent(BaseModel):
-    file_id: Optional[str] = None
-    """The ID of the file to be sent to the model."""
-
-    type: Literal["input_file"]
-    """The type of the input item. Always `input_file`."""
-
-    file_url: Optional[str] = None
-    """The URL of the file to be sent to the model."""
-
-    filename: Optional[str] = None
-    """The name of the file to be sent to the model."""
+InputFileContent = ResponseInputFile
diff --git a/src/openai/types/conversations/input_file_content_param.py b/src/openai/types/conversations/input_file_content_param.py
new file mode 100644
index 0000000000..1ed8b8b9d1
--- /dev/null
+++ b/src/openai/types/conversations/input_file_content_param.py
@@ -0,0 +1,7 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from ..responses.response_input_file_param import ResponseInputFileParam
+
+InputFileContentParam = ResponseInputFileParam
diff --git a/src/openai/types/conversations/input_image_content.py b/src/openai/types/conversations/input_image_content.py
index f2587e0adc..4304323c3a 100644
--- a/src/openai/types/conversations/input_image_content.py
+++ b/src/openai/types/conversations/input_image_content.py
@@ -1,28 +1,7 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
-from typing import Optional
-from typing_extensions import Literal
-
-from ..._models import BaseModel
+from ..responses.response_input_image import ResponseInputImage
 
 __all__ = ["InputImageContent"]
 
-
-class InputImageContent(BaseModel):
-    detail: Literal["low", "high", "auto"]
-    """The detail level of the image to be sent to the model.
-
-    One of `high`, `low`, or `auto`. Defaults to `auto`.
-    """
-
-    file_id: Optional[str] = None
-    """The ID of the file to be sent to the model."""
-
-    image_url: Optional[str] = None
-    """The URL of the image to be sent to the model.
-
-    A fully qualified URL or base64 encoded image in a data URL.
-    """
-
-    type: Literal["input_image"]
-    """The type of the input item. Always `input_image`."""
+InputImageContent = ResponseInputImage
diff --git a/src/openai/types/conversations/input_image_content_param.py b/src/openai/types/conversations/input_image_content_param.py
new file mode 100644
index 0000000000..a0ef9f545c
--- /dev/null
+++ b/src/openai/types/conversations/input_image_content_param.py
@@ -0,0 +1,7 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from ..responses.response_input_image_param import ResponseInputImageParam
+
+InputImageContentParam = ResponseInputImageParam
diff --git a/src/openai/types/conversations/input_text_content.py b/src/openai/types/conversations/input_text_content.py
index 5e2daebdc5..cab8b26cb1 100644
--- a/src/openai/types/conversations/input_text_content.py
+++ b/src/openai/types/conversations/input_text_content.py
@@ -1,15 +1,7 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
-from typing_extensions import Literal
-
-from ..._models import BaseModel
+from ..responses.response_input_text import ResponseInputText
 
 __all__ = ["InputTextContent"]
 
-
-class InputTextContent(BaseModel):
-    text: str
-    """The text input to the model."""
-
-    type: Literal["input_text"]
-    """The type of the input item. Always `input_text`."""
+InputTextContent = ResponseInputText
diff --git a/src/openai/types/conversations/input_text_content_param.py b/src/openai/types/conversations/input_text_content_param.py
new file mode 100644
index 0000000000..b1fd9a5f1c
--- /dev/null
+++ b/src/openai/types/conversations/input_text_content_param.py
@@ -0,0 +1,7 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from ..responses.response_input_text_param import ResponseInputTextParam
+
+InputTextContentParam = ResponseInputTextParam
diff --git a/src/openai/types/conversations/lob_prob.py b/src/openai/types/conversations/lob_prob.py
deleted file mode 100644
index f7dcd62a5e..0000000000
--- a/src/openai/types/conversations/lob_prob.py
+++ /dev/null
@@ -1,18 +0,0 @@
-# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
-
-from typing import List
-
-from ..._models import BaseModel
-from .top_log_prob import TopLogProb
-
-__all__ = ["LobProb"]
-
-
-class LobProb(BaseModel):
-    token: str
-
-    bytes: List[int]
-
-    logprob: float
-
-    top_logprobs: List[TopLogProb]
diff --git a/src/openai/types/conversations/message.py b/src/openai/types/conversations/message.py
index a070cf2869..95e03c5c00 100644
--- a/src/openai/types/conversations/message.py
+++ b/src/openai/types/conversations/message.py
@@ -6,26 +6,26 @@
 from ..._utils import PropertyInfo
 from ..._models import BaseModel
 from .text_content import TextContent
-from .refusal_content import RefusalContent
-from .input_file_content import InputFileContent
-from .input_text_content import InputTextContent
-from .input_image_content import InputImageContent
-from .output_text_content import OutputTextContent
 from .summary_text_content import SummaryTextContent
 from .computer_screenshot_content import ComputerScreenshotContent
+from ..responses.response_input_file import ResponseInputFile
+from ..responses.response_input_text import ResponseInputText
+from ..responses.response_input_image import ResponseInputImage
+from ..responses.response_output_text import ResponseOutputText
+from ..responses.response_output_refusal import ResponseOutputRefusal
 
 __all__ = ["Message", "Content"]
 
 Content: TypeAlias = Annotated[
     Union[
-        InputTextContent,
-        OutputTextContent,
+        ResponseInputText,
+        ResponseOutputText,
         TextContent,
         SummaryTextContent,
-        RefusalContent,
-        InputImageContent,
+        ResponseOutputRefusal,
+        ResponseInputImage,
         ComputerScreenshotContent,
-        InputFileContent,
+        ResponseInputFile,
     ],
     PropertyInfo(discriminator="type"),
 ]
diff --git a/src/openai/types/conversations/output_text_content.py b/src/openai/types/conversations/output_text_content.py
index 2ffee76526..cfe9307d74 100644
--- a/src/openai/types/conversations/output_text_content.py
+++ b/src/openai/types/conversations/output_text_content.py
@@ -1,30 +1,7 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
-from typing import List, Union, Optional
-from typing_extensions import Literal, Annotated, TypeAlias
+from ..responses.response_output_text import ResponseOutputText
 
-from ..._utils import PropertyInfo
-from .lob_prob import LobProb
-from ..._models import BaseModel
-from .url_citation_body import URLCitationBody
-from .file_citation_body import FileCitationBody
-from .container_file_citation_body import ContainerFileCitationBody
+__all__ = ["OutputTextContent"]
 
-__all__ = ["OutputTextContent", "Annotation"]
-
-Annotation: TypeAlias = Annotated[
-    Union[FileCitationBody, URLCitationBody, ContainerFileCitationBody], PropertyInfo(discriminator="type")
-]
-
-
-class OutputTextContent(BaseModel):
-    annotations: List[Annotation]
-    """The annotations of the text output."""
-
-    text: str
-    """The text output from the model."""
-
-    type: Literal["output_text"]
-    """The type of the output text. Always `output_text`."""
-
-    logprobs: Optional[List[LobProb]] = None
+OutputTextContent = ResponseOutputText
diff --git a/src/openai/types/conversations/output_text_content_param.py b/src/openai/types/conversations/output_text_content_param.py
new file mode 100644
index 0000000000..dc3e2026f6
--- /dev/null
+++ b/src/openai/types/conversations/output_text_content_param.py
@@ -0,0 +1,7 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from ..responses.response_output_text_param import ResponseOutputTextParam
+
+OutputTextContentParam = ResponseOutputTextParam
diff --git a/src/openai/types/conversations/refusal_content.py b/src/openai/types/conversations/refusal_content.py
index 3c8bd5e35f..6206c267dc 100644
--- a/src/openai/types/conversations/refusal_content.py
+++ b/src/openai/types/conversations/refusal_content.py
@@ -1,15 +1,7 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
-from typing_extensions import Literal
-
-from ..._models import BaseModel
+from ..responses.response_output_refusal import ResponseOutputRefusal
 
 __all__ = ["RefusalContent"]
 
-
-class RefusalContent(BaseModel):
-    refusal: str
-    """The refusal explanation from the model."""
-
-    type: Literal["refusal"]
-    """The type of the refusal. Always `refusal`."""
+RefusalContent = ResponseOutputRefusal
diff --git a/src/openai/types/conversations/refusal_content_param.py b/src/openai/types/conversations/refusal_content_param.py
new file mode 100644
index 0000000000..9b83da5f2d
--- /dev/null
+++ b/src/openai/types/conversations/refusal_content_param.py
@@ -0,0 +1,7 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from ..responses.response_output_refusal_param import ResponseOutputRefusalParam
+
+RefusalContentParam = ResponseOutputRefusalParam
diff --git a/src/openai/types/conversations/top_log_prob.py b/src/openai/types/conversations/top_log_prob.py
deleted file mode 100644
index fafca756ae..0000000000
--- a/src/openai/types/conversations/top_log_prob.py
+++ /dev/null
@@ -1,15 +0,0 @@
-# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
-
-from typing import List
-
-from ..._models import BaseModel
-
-__all__ = ["TopLogProb"]
-
-
-class TopLogProb(BaseModel):
-    token: str
-
-    bytes: List[int]
-
-    logprob: float
diff --git a/src/openai/types/conversations/url_citation_body.py b/src/openai/types/conversations/url_citation_body.py
deleted file mode 100644
index 1becb44bc0..0000000000
--- a/src/openai/types/conversations/url_citation_body.py
+++ /dev/null
@@ -1,24 +0,0 @@
-# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
-
-from typing_extensions import Literal
-
-from ..._models import BaseModel
-
-__all__ = ["URLCitationBody"]
-
-
-class URLCitationBody(BaseModel):
-    end_index: int
-    """The index of the last character of the URL citation in the message."""
-
-    start_index: int
-    """The index of the first character of the URL citation in the message."""
-
-    title: str
-    """The title of the web resource."""
-
-    type: Literal["url_citation"]
-    """The type of the URL citation. Always `url_citation`."""
-
-    url: str
-    """The URL of the web resource."""
diff --git a/src/openai/types/evals/create_eval_completions_run_data_source.py b/src/openai/types/evals/create_eval_completions_run_data_source.py
index edf70c8ad4..74323a735e 100644
--- a/src/openai/types/evals/create_eval_completions_run_data_source.py
+++ b/src/openai/types/evals/create_eval_completions_run_data_source.py
@@ -6,6 +6,7 @@
 from ..._utils import PropertyInfo
 from ..._models import BaseModel
 from ..shared.metadata import Metadata
+from ..shared.reasoning_effort import ReasoningEffort
 from ..shared.response_format_text import ResponseFormatText
 from ..responses.easy_input_message import EasyInputMessage
 from ..responses.response_input_text import ResponseInputText
@@ -167,6 +168,15 @@ class SamplingParams(BaseModel):
     max_completion_tokens: Optional[int] = None
     """The maximum number of tokens in the generated output."""
 
+    reasoning_effort: Optional[ReasoningEffort] = None
+    """
+    Constrains effort on reasoning for
+    [reasoning models](https://platform.openai.com/docs/guides/reasoning). Currently
+    supported values are `minimal`, `low`, `medium`, and `high`. Reducing reasoning
+    effort can result in faster responses and fewer tokens used on reasoning in a
+    response.
+    """
+
     response_format: Optional[SamplingParamsResponseFormat] = None
     """An object specifying the format that the model must output.
 
diff --git a/src/openai/types/evals/create_eval_completions_run_data_source_param.py b/src/openai/types/evals/create_eval_completions_run_data_source_param.py
index c14360ac80..4e9c1fdeb8 100644
--- a/src/openai/types/evals/create_eval_completions_run_data_source_param.py
+++ b/src/openai/types/evals/create_eval_completions_run_data_source_param.py
@@ -6,6 +6,7 @@
 from typing_extensions import Literal, Required, TypeAlias, TypedDict
 
 from ..shared_params.metadata import Metadata
+from ..shared.reasoning_effort import ReasoningEffort
 from ..responses.easy_input_message_param import EasyInputMessageParam
 from ..shared_params.response_format_text import ResponseFormatText
 from ..responses.response_input_text_param import ResponseInputTextParam
@@ -163,6 +164,15 @@ class SamplingParams(TypedDict, total=False):
     max_completion_tokens: int
     """The maximum number of tokens in the generated output."""
 
+    reasoning_effort: Optional[ReasoningEffort]
+    """
+    Constrains effort on reasoning for
+    [reasoning models](https://platform.openai.com/docs/guides/reasoning). Currently
+    supported values are `minimal`, `low`, `medium`, and `high`. Reducing reasoning
+    effort can result in faster responses and fewer tokens used on reasoning in a
+    response.
+    """
+
     response_format: SamplingParamsResponseFormat
     """An object specifying the format that the model must output.
 
diff --git a/src/openai/types/evals/run_cancel_response.py b/src/openai/types/evals/run_cancel_response.py
index 8f43494e68..d04d4ff657 100644
--- a/src/openai/types/evals/run_cancel_response.py
+++ b/src/openai/types/evals/run_cancel_response.py
@@ -234,6 +234,15 @@ class DataSourceResponsesSamplingParams(BaseModel):
     max_completion_tokens: Optional[int] = None
     """The maximum number of tokens in the generated output."""
 
+    reasoning_effort: Optional[ReasoningEffort] = None
+    """
+    Constrains effort on reasoning for
+    [reasoning models](https://platform.openai.com/docs/guides/reasoning). Currently
+    supported values are `minimal`, `low`, `medium`, and `high`. Reducing reasoning
+    effort can result in faster responses and fewer tokens used on reasoning in a
+    response.
+    """
+
     seed: Optional[int] = None
     """A seed value to initialize the randomness, during sampling."""
 
diff --git a/src/openai/types/evals/run_create_params.py b/src/openai/types/evals/run_create_params.py
index 35813c8901..6ff897b5de 100644
--- a/src/openai/types/evals/run_create_params.py
+++ b/src/openai/types/evals/run_create_params.py
@@ -252,6 +252,15 @@ class DataSourceCreateEvalResponsesRunDataSourceSamplingParams(TypedDict, total=
     max_completion_tokens: int
     """The maximum number of tokens in the generated output."""
 
+    reasoning_effort: Optional[ReasoningEffort]
+    """
+    Constrains effort on reasoning for
+    [reasoning models](https://platform.openai.com/docs/guides/reasoning). Currently
+    supported values are `minimal`, `low`, `medium`, and `high`. Reducing reasoning
+    effort can result in faster responses and fewer tokens used on reasoning in a
+    response.
+    """
+
     seed: int
     """A seed value to initialize the randomness, during sampling."""
 
diff --git a/src/openai/types/evals/run_create_response.py b/src/openai/types/evals/run_create_response.py
index c842a5ad2f..defa275c8c 100644
--- a/src/openai/types/evals/run_create_response.py
+++ b/src/openai/types/evals/run_create_response.py
@@ -234,6 +234,15 @@ class DataSourceResponsesSamplingParams(BaseModel):
     max_completion_tokens: Optional[int] = None
     """The maximum number of tokens in the generated output."""
 
+    reasoning_effort: Optional[ReasoningEffort] = None
+    """
+    Constrains effort on reasoning for
+    [reasoning models](https://platform.openai.com/docs/guides/reasoning). Currently
+    supported values are `minimal`, `low`, `medium`, and `high`. Reducing reasoning
+    effort can result in faster responses and fewer tokens used on reasoning in a
+    response.
+    """
+
     seed: Optional[int] = None
     """A seed value to initialize the randomness, during sampling."""
 
diff --git a/src/openai/types/evals/run_list_response.py b/src/openai/types/evals/run_list_response.py
index 5a5c2efbb3..7fe0e55ace 100644
--- a/src/openai/types/evals/run_list_response.py
+++ b/src/openai/types/evals/run_list_response.py
@@ -234,6 +234,15 @@ class DataSourceResponsesSamplingParams(BaseModel):
     max_completion_tokens: Optional[int] = None
     """The maximum number of tokens in the generated output."""
 
+    reasoning_effort: Optional[ReasoningEffort] = None
+    """
+    Constrains effort on reasoning for
+    [reasoning models](https://platform.openai.com/docs/guides/reasoning). Currently
+    supported values are `minimal`, `low`, `medium`, and `high`. Reducing reasoning
+    effort can result in faster responses and fewer tokens used on reasoning in a
+    response.
+    """
+
     seed: Optional[int] = None
     """A seed value to initialize the randomness, during sampling."""
 
diff --git a/src/openai/types/evals/run_retrieve_response.py b/src/openai/types/evals/run_retrieve_response.py
index f341296875..a50520f17d 100644
--- a/src/openai/types/evals/run_retrieve_response.py
+++ b/src/openai/types/evals/run_retrieve_response.py
@@ -234,6 +234,15 @@ class DataSourceResponsesSamplingParams(BaseModel):
     max_completion_tokens: Optional[int] = None
     """The maximum number of tokens in the generated output."""
 
+    reasoning_effort: Optional[ReasoningEffort] = None
+    """
+    Constrains effort on reasoning for
+    [reasoning models](https://platform.openai.com/docs/guides/reasoning). Currently
+    supported values are `minimal`, `low`, `medium`, and `high`. Reducing reasoning
+    effort can result in faster responses and fewer tokens used on reasoning in a
+    response.
+    """
+
     seed: Optional[int] = None
     """A seed value to initialize the randomness, during sampling."""
 
diff --git a/src/openai/types/evals/runs/output_item_list_response.py b/src/openai/types/evals/runs/output_item_list_response.py
index 72b1049f7b..f774518f3c 100644
--- a/src/openai/types/evals/runs/output_item_list_response.py
+++ b/src/openai/types/evals/runs/output_item_list_response.py
@@ -1,13 +1,38 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
-import builtins
-from typing import Dict, List, Optional
+from typing import TYPE_CHECKING, Dict, List, Optional
 from typing_extensions import Literal
 
+from pydantic import Field as FieldInfo
+
 from ...._models import BaseModel
 from ..eval_api_error import EvalAPIError
 
-__all__ = ["OutputItemListResponse", "Sample", "SampleInput", "SampleOutput", "SampleUsage"]
+__all__ = ["OutputItemListResponse", "Result", "Sample", "SampleInput", "SampleOutput", "SampleUsage"]
+
+
+class Result(BaseModel):
+    name: str
+    """The name of the grader."""
+
+    passed: bool
+    """Whether the grader considered the output a pass."""
+
+    score: float
+    """The numeric score produced by the grader."""
+
+    sample: Optional[Dict[str, object]] = None
+    """Optional sample or intermediate data produced by the grader."""
+
+    type: Optional[str] = None
+    """The grader type (for example, "string-check-grader")."""
+
+    __pydantic_extra__: Dict[str, object] = FieldInfo(init=False)  # pyright: ignore[reportIncompatibleVariableOverride]
+    if TYPE_CHECKING:
+        # Stub to indicate that arbitrary properties are accepted.
+        # To access properties that are not valid identifiers you can use `getattr`, e.g.
+        # `getattr(obj, '$type')`
+        def __getattr__(self, attr: str) -> object: ...
 
 
 class SampleInput(BaseModel):
@@ -91,8 +116,8 @@ class OutputItemListResponse(BaseModel):
     object: Literal["eval.run.output_item"]
     """The type of the object. Always "eval.run.output_item"."""
 
-    results: List[Dict[str, builtins.object]]
-    """A list of results from the evaluation run."""
+    results: List[Result]
+    """A list of grader results for this output item."""
 
     run_id: str
     """The identifier of the evaluation run associated with this output item."""
diff --git a/src/openai/types/evals/runs/output_item_retrieve_response.py b/src/openai/types/evals/runs/output_item_retrieve_response.py
index 63aab5565f..d66435bd4f 100644
--- a/src/openai/types/evals/runs/output_item_retrieve_response.py
+++ b/src/openai/types/evals/runs/output_item_retrieve_response.py
@@ -1,13 +1,38 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
-import builtins
-from typing import Dict, List, Optional
+from typing import TYPE_CHECKING, Dict, List, Optional
 from typing_extensions import Literal
 
+from pydantic import Field as FieldInfo
+
 from ...._models import BaseModel
 from ..eval_api_error import EvalAPIError
 
-__all__ = ["OutputItemRetrieveResponse", "Sample", "SampleInput", "SampleOutput", "SampleUsage"]
+__all__ = ["OutputItemRetrieveResponse", "Result", "Sample", "SampleInput", "SampleOutput", "SampleUsage"]
+
+
+class Result(BaseModel):
+    name: str
+    """The name of the grader."""
+
+    passed: bool
+    """Whether the grader considered the output a pass."""
+
+    score: float
+    """The numeric score produced by the grader."""
+
+    sample: Optional[Dict[str, object]] = None
+    """Optional sample or intermediate data produced by the grader."""
+
+    type: Optional[str] = None
+    """The grader type (for example, "string-check-grader")."""
+
+    __pydantic_extra__: Dict[str, object] = FieldInfo(init=False)  # pyright: ignore[reportIncompatibleVariableOverride]
+    if TYPE_CHECKING:
+        # Stub to indicate that arbitrary properties are accepted.
+        # To access properties that are not valid identifiers you can use `getattr`, e.g.
+        # `getattr(obj, '$type')`
+        def __getattr__(self, attr: str) -> object: ...
 
 
 class SampleInput(BaseModel):
@@ -91,8 +116,8 @@ class OutputItemRetrieveResponse(BaseModel):
     object: Literal["eval.run.output_item"]
     """The type of the object. Always "eval.run.output_item"."""
 
-    results: List[Dict[str, builtins.object]]
-    """A list of results from the evaluation run."""
+    results: List[Result]
+    """A list of grader results for this output item."""
 
     run_id: str
     """The identifier of the evaluation run associated with this output item."""
diff --git a/src/openai/types/graders/score_model_grader.py b/src/openai/types/graders/score_model_grader.py
index fc221b8e41..908c6f91d3 100644
--- a/src/openai/types/graders/score_model_grader.py
+++ b/src/openai/types/graders/score_model_grader.py
@@ -4,10 +4,18 @@
 from typing_extensions import Literal, TypeAlias
 
 from ..._models import BaseModel
+from ..shared.reasoning_effort import ReasoningEffort
 from ..responses.response_input_text import ResponseInputText
 from ..responses.response_input_audio import ResponseInputAudio
 
-__all__ = ["ScoreModelGrader", "Input", "InputContent", "InputContentOutputText", "InputContentInputImage"]
+__all__ = [
+    "ScoreModelGrader",
+    "Input",
+    "InputContent",
+    "InputContentOutputText",
+    "InputContentInputImage",
+    "SamplingParams",
+]
 
 
 class InputContentOutputText(BaseModel):
@@ -51,6 +59,29 @@ class Input(BaseModel):
     """The type of the message input. Always `message`."""
 
 
+class SamplingParams(BaseModel):
+    max_completions_tokens: Optional[int] = None
+    """The maximum number of tokens the grader model may generate in its response."""
+
+    reasoning_effort: Optional[ReasoningEffort] = None
+    """
+    Constrains effort on reasoning for
+    [reasoning models](https://platform.openai.com/docs/guides/reasoning). Currently
+    supported values are `minimal`, `low`, `medium`, and `high`. Reducing reasoning
+    effort can result in faster responses and fewer tokens used on reasoning in a
+    response.
+    """
+
+    seed: Optional[int] = None
+    """A seed value to initialize the randomness, during sampling."""
+
+    temperature: Optional[float] = None
+    """A higher temperature increases randomness in the outputs."""
+
+    top_p: Optional[float] = None
+    """An alternative to temperature for nucleus sampling; 1.0 includes all tokens."""
+
+
 class ScoreModelGrader(BaseModel):
     input: List[Input]
     """The input text. This may include template strings."""
@@ -67,5 +98,5 @@ class ScoreModelGrader(BaseModel):
     range: Optional[List[float]] = None
     """The range of the score. Defaults to `[0, 1]`."""
 
-    sampling_params: Optional[object] = None
+    sampling_params: Optional[SamplingParams] = None
     """The sampling parameters for the model."""
diff --git a/src/openai/types/graders/score_model_grader_param.py b/src/openai/types/graders/score_model_grader_param.py
index 15100bb74b..743944e099 100644
--- a/src/openai/types/graders/score_model_grader_param.py
+++ b/src/openai/types/graders/score_model_grader_param.py
@@ -2,13 +2,21 @@
 
 from __future__ import annotations
 
-from typing import Union, Iterable
+from typing import Union, Iterable, Optional
 from typing_extensions import Literal, Required, TypeAlias, TypedDict
 
+from ..shared.reasoning_effort import ReasoningEffort
 from ..responses.response_input_text_param import ResponseInputTextParam
 from ..responses.response_input_audio_param import ResponseInputAudioParam
 
-__all__ = ["ScoreModelGraderParam", "Input", "InputContent", "InputContentOutputText", "InputContentInputImage"]
+__all__ = [
+    "ScoreModelGraderParam",
+    "Input",
+    "InputContent",
+    "InputContentOutputText",
+    "InputContentInputImage",
+    "SamplingParams",
+]
 
 
 class InputContentOutputText(TypedDict, total=False):
@@ -57,6 +65,29 @@ class Input(TypedDict, total=False):
     """The type of the message input. Always `message`."""
 
 
+class SamplingParams(TypedDict, total=False):
+    max_completions_tokens: Optional[int]
+    """The maximum number of tokens the grader model may generate in its response."""
+
+    reasoning_effort: Optional[ReasoningEffort]
+    """
+    Constrains effort on reasoning for
+    [reasoning models](https://platform.openai.com/docs/guides/reasoning). Currently
+    supported values are `minimal`, `low`, `medium`, and `high`. Reducing reasoning
+    effort can result in faster responses and fewer tokens used on reasoning in a
+    response.
+    """
+
+    seed: Optional[int]
+    """A seed value to initialize the randomness, during sampling."""
+
+    temperature: Optional[float]
+    """A higher temperature increases randomness in the outputs."""
+
+    top_p: Optional[float]
+    """An alternative to temperature for nucleus sampling; 1.0 includes all tokens."""
+
+
 class ScoreModelGraderParam(TypedDict, total=False):
     input: Required[Iterable[Input]]
     """The input text. This may include template strings."""
@@ -73,5 +104,5 @@ class ScoreModelGraderParam(TypedDict, total=False):
     range: Iterable[float]
     """The range of the score. Defaults to `[0, 1]`."""
 
-    sampling_params: object
+    sampling_params: SamplingParams
     """The sampling parameters for the model."""