diff --git a/.release-please-manifest.json b/.release-please-manifest.json
index 804a6039aa..5d9ceab581 100644
--- a/.release-please-manifest.json
+++ b/.release-please-manifest.json
@@ -1,3 +1,3 @@
 {
-  ".": "1.99.7"
+  ".": "1.99.8"
 }
\ No newline at end of file
diff --git a/CHANGELOG.md b/CHANGELOG.md
index 74d0da964a..33e0e8e948 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,5 +1,26 @@
 # Changelog
 
+## 1.99.8 (2025-08-11)
+
+Full Changelog: [v1.99.7...v1.99.8](https://github.com/openai/openai-python/compare/v1.99.7...v1.99.8)
+
+### Bug Fixes
+
+* **internal/tests:** correct snapshot update comment ([2784a7a](https://github.com/openai/openai-python/commit/2784a7a7da24ddba74b5717f07d67546864472b9))
+* **types:** revert ChatCompletionMessageToolCallUnion breaking change ([ba54e03](https://github.com/openai/openai-python/commit/ba54e03bc2d21825d891685bf3bad4a9253cbeb0))
+
+
+### Chores
+
+* **internal/tests:** add inline snapshot format command ([8107db8](https://github.com/openai/openai-python/commit/8107db8ff738baa65fe4cf2f2d7f1acd29219c78))
+* **internal:** fix formatting ([f03a03d](https://github.com/openai/openai-python/commit/f03a03de8c84740209d021598ff8bf56b6d3c684))
+* **tests:** add responses output_text test ([971347b](https://github.com/openai/openai-python/commit/971347b3a05f79c51abd11c86b382ca73c28cefb))
+
+
+### Refactors
+
+* **tests:** share snapshot utils ([791c567](https://github.com/openai/openai-python/commit/791c567cd87fb8d587965773b1da0404c7848c68))
+
 ## 1.99.7 (2025-08-11)
 
 Full Changelog: [v1.99.6...v1.99.7](https://github.com/openai/openai-python/compare/v1.99.6...v1.99.7)
diff --git a/pyproject.toml b/pyproject.toml
index d58b9b1eb2..b4a7d01a2b 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -1,6 +1,6 @@
 [project]
 name = "openai"
-version = "1.99.7"
+version = "1.99.8"
 description = "The official Python library for the openai API"
 dynamic = ["readme"]
 license = "Apache-2.0"
@@ -150,6 +150,9 @@ filterwarnings = [
   "error"
 ]
 
+[tool.inline-snapshot]
+format-command="ruff format --stdin-filename {filename}"
+
 [tool.pyright]
 # this enables practically every flag given by pyright.
 # there are a couple of flags that are still disabled by
diff --git a/src/openai/_version.py b/src/openai/_version.py
index 3db3f866cf..9d1f1f4e96 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__ = "1.99.7"  # x-release-please-version
+__version__ = "1.99.8"  # x-release-please-version
diff --git a/src/openai/types/chat/__init__.py b/src/openai/types/chat/__init__.py
index 2aecaf7d0c..50bdac7c65 100644
--- a/src/openai/types/chat/__init__.py
+++ b/src/openai/types/chat/__init__.py
@@ -31,7 +31,10 @@
 from .chat_completion_tool_union_param import ChatCompletionToolUnionParam as ChatCompletionToolUnionParam
 from .chat_completion_content_part_text import ChatCompletionContentPartText as ChatCompletionContentPartText
 from .chat_completion_custom_tool_param import ChatCompletionCustomToolParam as ChatCompletionCustomToolParam
-from .chat_completion_message_tool_call import ChatCompletionMessageToolCallUnion as ChatCompletionMessageToolCallUnion
+from .chat_completion_message_tool_call import (
+    ChatCompletionMessageToolCall as ChatCompletionMessageToolCall,
+    ChatCompletionMessageToolCallUnion as ChatCompletionMessageToolCallUnion,
+)
 from .chat_completion_content_part_image import ChatCompletionContentPartImage as ChatCompletionContentPartImage
 from .chat_completion_content_part_param import ChatCompletionContentPartParam as ChatCompletionContentPartParam
 from .chat_completion_tool_message_param import ChatCompletionToolMessageParam as ChatCompletionToolMessageParam
diff --git a/src/openai/types/chat/chat_completion_message_tool_call.py b/src/openai/types/chat/chat_completion_message_tool_call.py
index df687b19bd..845e639089 100644
--- a/src/openai/types/chat/chat_completion_message_tool_call.py
+++ b/src/openai/types/chat/chat_completion_message_tool_call.py
@@ -7,9 +7,11 @@
 from .chat_completion_message_custom_tool_call import ChatCompletionMessageCustomToolCall
 from .chat_completion_message_function_tool_call import Function as Function, ChatCompletionMessageFunctionToolCall
 
-__all__ = [ "Function", "ChatCompletionMessageToolCallUnion"]
+__all__ = ["Function", "ChatCompletionMessageToolCallUnion"]
 
 ChatCompletionMessageToolCallUnion: TypeAlias = Annotated[
     Union[ChatCompletionMessageFunctionToolCall, ChatCompletionMessageCustomToolCall],
     PropertyInfo(discriminator="type"),
 ]
+
+ChatCompletionMessageToolCall: TypeAlias = ChatCompletionMessageToolCallUnion
diff --git a/tests/lib/chat/test_completions.py b/tests/lib/chat/test_completions.py
index d0bd14ce9e..f04a0e3782 100644
--- a/tests/lib/chat/test_completions.py
+++ b/tests/lib/chat/test_completions.py
@@ -1,12 +1,9 @@
 from __future__ import annotations
 
-import os
-import json
 from enum import Enum
-from typing import Any, List, Callable, Optional, Awaitable
+from typing import List, Optional
 from typing_extensions import Literal, TypeVar
 
-import httpx
 import pytest
 from respx import MockRouter
 from pydantic import Field, BaseModel
@@ -17,8 +14,9 @@
 from openai._utils import assert_signatures_in_sync
 from openai._compat import PYDANTIC_V2
 
-from ._utils import print_obj, get_snapshot_value
+from ..utils import print_obj
 from ...conftest import base_url
+from ..snapshots import make_snapshot_request, make_async_snapshot_request
 from ..schema_types.query import Query
 
 _T = TypeVar("_T")
@@ -27,12 +25,12 @@
 #
 # you can update them with
 #
-# `OPENAI_LIVE=1 pytest --inline-snapshot=fix`
+# `OPENAI_LIVE=1 pytest --inline-snapshot=fix -p no:xdist -o addopts=""`
 
 
 @pytest.mark.respx(base_url=base_url)
 def test_parse_nothing(client: OpenAI, respx_mock: MockRouter, monkeypatch: pytest.MonkeyPatch) -> None:
-    completion = _make_snapshot_request(
+    completion = make_snapshot_request(
         lambda c: c.chat.completions.parse(
             model="gpt-4o-2024-08-06",
             messages=[
@@ -45,6 +43,7 @@ def test_parse_nothing(client: OpenAI, respx_mock: MockRouter, monkeypatch: pyte
         content_snapshot=snapshot(
             '{"id": "chatcmpl-ABfvaueLEMLNYbT8YzpJxsmiQ6HSY", "object": "chat.completion", "created": 1727346142, "model": "gpt-4o-2024-08-06", "choices": [{"index": 0, "message": {"role": "assistant", "content": "I\'m unable to provide real-time weather updates. To get the current weather in San Francisco, I recommend checking a reliable weather website or app like the Weather Channel or a local news station.", "refusal": null}, "logprobs": null, "finish_reason": "stop"}], "usage": {"prompt_tokens": 14, "completion_tokens": 37, "total_tokens": 51, "completion_tokens_details": {"reasoning_tokens": 0}}, "system_fingerprint": "fp_b40fb1c6fb"}'
         ),
+        path="/chat/completions",
         mock_client=client,
         respx_mock=respx_mock,
     )
@@ -100,7 +99,7 @@ class Location(BaseModel):
         temperature: float
         units: Literal["c", "f"]
 
-    completion = _make_snapshot_request(
+    completion = make_snapshot_request(
         lambda c: c.chat.completions.parse(
             model="gpt-4o-2024-08-06",
             messages=[
@@ -114,6 +113,7 @@ class Location(BaseModel):
         content_snapshot=snapshot(
             '{"id": "chatcmpl-ABfvbtVnTu5DeC4EFnRYj8mtfOM99", "object": "chat.completion", "created": 1727346143, "model": "gpt-4o-2024-08-06", "choices": [{"index": 0, "message": {"role": "assistant", "content": "{\\"city\\":\\"San Francisco\\",\\"temperature\\":65,\\"units\\":\\"f\\"}", "refusal": null}, "logprobs": null, "finish_reason": "stop"}], "usage": {"prompt_tokens": 79, "completion_tokens": 14, "total_tokens": 93, "completion_tokens_details": {"reasoning_tokens": 0}}, "system_fingerprint": "fp_5050236cbd"}'
         ),
+        path="/chat/completions",
         mock_client=client,
         respx_mock=respx_mock,
     )
@@ -170,7 +170,7 @@ class Location(BaseModel):
         temperature: float
         units: Optional[Literal["c", "f"]] = None
 
-    completion = _make_snapshot_request(
+    completion = make_snapshot_request(
         lambda c: c.chat.completions.parse(
             model="gpt-4o-2024-08-06",
             messages=[
@@ -184,6 +184,7 @@ class Location(BaseModel):
         content_snapshot=snapshot(
             '{"id": "chatcmpl-ABfvcC8grKYsRkSoMp9CCAhbXAd0b", "object": "chat.completion", "created": 1727346144, "model": "gpt-4o-2024-08-06", "choices": [{"index": 0, "message": {"role": "assistant", "content": "{\\"city\\":\\"San Francisco\\",\\"temperature\\":65,\\"units\\":\\"f\\"}", "refusal": null}, "logprobs": null, "finish_reason": "stop"}], "usage": {"prompt_tokens": 88, "completion_tokens": 14, "total_tokens": 102, "completion_tokens_details": {"reasoning_tokens": 0}}, "system_fingerprint": "fp_b40fb1c6fb"}'
         ),
+        path="/chat/completions",
         mock_client=client,
         respx_mock=respx_mock,
     )
@@ -247,7 +248,7 @@ class ColorDetection(BaseModel):
     if not PYDANTIC_V2:
         ColorDetection.update_forward_refs(**locals())  # type: ignore
 
-    completion = _make_snapshot_request(
+    completion = make_snapshot_request(
         lambda c: c.chat.completions.parse(
             model="gpt-4o-2024-08-06",
             messages=[
@@ -258,6 +259,7 @@ class ColorDetection(BaseModel):
         content_snapshot=snapshot(
             '{"id": "chatcmpl-ABfvjIatz0zrZu50gRbMtlp0asZpz", "object": "chat.completion", "created": 1727346151, "model": "gpt-4o-2024-08-06", "choices": [{"index": 0, "message": {"role": "assistant", "content": "{\\"color\\":\\"red\\",\\"hex_color_code\\":\\"#FF0000\\"}", "refusal": null}, "logprobs": null, "finish_reason": "stop"}], "usage": {"prompt_tokens": 109, "completion_tokens": 14, "total_tokens": 123, "completion_tokens_details": {"reasoning_tokens": 0}}, "system_fingerprint": "fp_5050236cbd"}'
         ),
+        path="/chat/completions",
         mock_client=client,
         respx_mock=respx_mock,
     )
@@ -292,7 +294,7 @@ class Location(BaseModel):
         temperature: float
         units: Literal["c", "f"]
 
-    completion = _make_snapshot_request(
+    completion = make_snapshot_request(
         lambda c: c.chat.completions.parse(
             model="gpt-4o-2024-08-06",
             messages=[
@@ -307,6 +309,7 @@ class Location(BaseModel):
         content_snapshot=snapshot(
             '{"id": "chatcmpl-ABfvp8qzboW92q8ONDF4DPHlI7ckC", "object": "chat.completion", "created": 1727346157, "model": "gpt-4o-2024-08-06", "choices": [{"index": 0, "message": {"role": "assistant", "content": "{\\"city\\":\\"San Francisco\\",\\"temperature\\":64,\\"units\\":\\"f\\"}", "refusal": null}, "logprobs": null, "finish_reason": "stop"}, {"index": 1, "message": {"role": "assistant", "content": "{\\"city\\":\\"San Francisco\\",\\"temperature\\":65,\\"units\\":\\"f\\"}", "refusal": null}, "logprobs": null, "finish_reason": "stop"}, {"index": 2, "message": {"role": "assistant", "content": "{\\"city\\":\\"San Francisco\\",\\"temperature\\":63.0,\\"units\\":\\"f\\"}", "refusal": null}, "logprobs": null, "finish_reason": "stop"}], "usage": {"prompt_tokens": 79, "completion_tokens": 44, "total_tokens": 123, "completion_tokens_details": {"reasoning_tokens": 0}}, "system_fingerprint": "fp_b40fb1c6fb"}'
         ),
+        path="/chat/completions",
         mock_client=client,
         respx_mock=respx_mock,
     )
@@ -375,7 +378,7 @@ class CalendarEvent:
         date: str
         participants: List[str]
 
-    completion = _make_snapshot_request(
+    completion = make_snapshot_request(
         lambda c: c.chat.completions.parse(
             model="gpt-4o-2024-08-06",
             messages=[
@@ -387,6 +390,7 @@ class CalendarEvent:
         content_snapshot=snapshot(
             '{"id": "chatcmpl-ABfvqhz4uUUWsw8Ohw2Mp9B4sKKV8", "object": "chat.completion", "created": 1727346158, "model": "gpt-4o-2024-08-06", "choices": [{"index": 0, "message": {"role": "assistant", "content": "{\\"name\\":\\"Science Fair\\",\\"date\\":\\"Friday\\",\\"participants\\":[\\"Alice\\",\\"Bob\\"]}", "refusal": null}, "logprobs": null, "finish_reason": "stop"}], "usage": {"prompt_tokens": 92, "completion_tokens": 17, "total_tokens": 109, "completion_tokens_details": {"reasoning_tokens": 0}}, "system_fingerprint": "fp_7568d46099"}'
         ),
+        path="/chat/completions",
         mock_client=client,
         respx_mock=respx_mock,
     )
@@ -436,7 +440,7 @@ class CalendarEvent:
 
 @pytest.mark.respx(base_url=base_url)
 def test_pydantic_tool_model_all_types(client: OpenAI, respx_mock: MockRouter, monkeypatch: pytest.MonkeyPatch) -> None:
-    completion = _make_snapshot_request(
+    completion = make_snapshot_request(
         lambda c: c.chat.completions.parse(
             model="gpt-4o-2024-08-06",
             messages=[
@@ -451,6 +455,7 @@ def test_pydantic_tool_model_all_types(client: OpenAI, respx_mock: MockRouter, m
         content_snapshot=snapshot(
             '{"id": "chatcmpl-ABfvtNiaTNUF6OymZUnEFc9lPq9p1", "object": "chat.completion", "created": 1727346161, "model": "gpt-4o-2024-08-06", "choices": [{"index": 0, "message": {"role": "assistant", "content": null, "tool_calls": [{"id": "call_NKpApJybW1MzOjZO2FzwYw0d", "type": "function", "function": {"name": "Query", "arguments": "{\\"name\\":\\"May 2022 Fulfilled Orders Not Delivered on Time\\",\\"table_name\\":\\"orders\\",\\"columns\\":[\\"id\\",\\"status\\",\\"expected_delivery_date\\",\\"delivered_at\\",\\"shipped_at\\",\\"ordered_at\\",\\"canceled_at\\"],\\"conditions\\":[{\\"column\\":\\"ordered_at\\",\\"operator\\":\\">=\\",\\"value\\":\\"2022-05-01\\"},{\\"column\\":\\"ordered_at\\",\\"operator\\":\\"<=\\",\\"value\\":\\"2022-05-31\\"},{\\"column\\":\\"status\\",\\"operator\\":\\"=\\",\\"value\\":\\"fulfilled\\"},{\\"column\\":\\"delivered_at\\",\\"operator\\":\\">\\",\\"value\\":{\\"column_name\\":\\"expected_delivery_date\\"}}],\\"order_by\\":\\"asc\\"}"}}], "refusal": null}, "logprobs": null, "finish_reason": "tool_calls"}], "usage": {"prompt_tokens": 512, "completion_tokens": 132, "total_tokens": 644, "completion_tokens_details": {"reasoning_tokens": 0}}, "system_fingerprint": "fp_7568d46099"}'
         ),
+        path="/chat/completions",
         mock_client=client,
         respx_mock=respx_mock,
     )
@@ -521,7 +526,7 @@ class Location(BaseModel):
         units: Literal["c", "f"]
 
     with pytest.raises(openai.LengthFinishReasonError):
-        _make_snapshot_request(
+        make_snapshot_request(
             lambda c: c.chat.completions.parse(
                 model="gpt-4o-2024-08-06",
                 messages=[
@@ -536,6 +541,7 @@ class Location(BaseModel):
             content_snapshot=snapshot(
                 '{"id": "chatcmpl-ABfvvX7eB1KsfeZj8VcF3z7G7SbaA", "object": "chat.completion", "created": 1727346163, "model": "gpt-4o-2024-08-06", "choices": [{"index": 0, "message": {"role": "assistant", "content": "{\\"", "refusal": null}, "logprobs": null, "finish_reason": "length"}], "usage": {"prompt_tokens": 79, "completion_tokens": 1, "total_tokens": 80, "completion_tokens_details": {"reasoning_tokens": 0}}, "system_fingerprint": "fp_7568d46099"}'
             ),
+        path="/chat/completions",
             mock_client=client,
             respx_mock=respx_mock,
         )
@@ -548,7 +554,7 @@ class Location(BaseModel):
         temperature: float
         units: Literal["c", "f"]
 
-    completion = _make_snapshot_request(
+    completion = make_snapshot_request(
         lambda c: c.chat.completions.parse(
             model="gpt-4o-2024-08-06",
             messages=[
@@ -562,6 +568,7 @@ class Location(BaseModel):
         content_snapshot=snapshot(
             '{"id": "chatcmpl-ABfvwoKVWPQj2UPlAcAKM7s40GsRx", "object": "chat.completion", "created": 1727346164, "model": "gpt-4o-2024-08-06", "choices": [{"index": 0, "message": {"role": "assistant", "content": null, "refusal": "I\'m very sorry, but I can\'t assist with that."}, "logprobs": null, "finish_reason": "stop"}], "usage": {"prompt_tokens": 79, "completion_tokens": 12, "total_tokens": 91, "completion_tokens_details": {"reasoning_tokens": 0}}, "system_fingerprint": "fp_5050236cbd"}'
         ),
+        path="/chat/completions",
         mock_client=client,
         respx_mock=respx_mock,
     )
@@ -596,7 +603,7 @@ class GetWeatherArgs(BaseModel):
         country: str
         units: Literal["c", "f"] = "c"
 
-    completion = _make_snapshot_request(
+    completion = make_snapshot_request(
         lambda c: c.chat.completions.parse(
             model="gpt-4o-2024-08-06",
             messages=[
@@ -612,6 +619,7 @@ class GetWeatherArgs(BaseModel):
         content_snapshot=snapshot(
             '{"id": "chatcmpl-ABfvx6Z4dchiW2nya1N8KMsHFrQRE", "object": "chat.completion", "created": 1727346165, "model": "gpt-4o-2024-08-06", "choices": [{"index": 0, "message": {"role": "assistant", "content": null, "tool_calls": [{"id": "call_Y6qJ7ofLgOrBnMD5WbVAeiRV", "type": "function", "function": {"name": "GetWeatherArgs", "arguments": "{\\"city\\":\\"Edinburgh\\",\\"country\\":\\"UK\\",\\"units\\":\\"c\\"}"}}], "refusal": null}, "logprobs": null, "finish_reason": "tool_calls"}], "usage": {"prompt_tokens": 76, "completion_tokens": 24, "total_tokens": 100, "completion_tokens_details": {"reasoning_tokens": 0}}, "system_fingerprint": "fp_e45dabd248"}'
         ),
+        path="/chat/completions",
         mock_client=client,
         respx_mock=respx_mock,
     )
@@ -662,7 +670,7 @@ class GetStockPrice(BaseModel):
         ticker: str
         exchange: str
 
-    completion = _make_snapshot_request(
+    completion = make_snapshot_request(
         lambda c: c.chat.completions.parse(
             model="gpt-4o-2024-08-06",
             messages=[
@@ -685,6 +693,7 @@ class GetStockPrice(BaseModel):
         content_snapshot=snapshot(
             '{"id": "chatcmpl-ABfvyvfNWKcl7Ohqos4UFrmMs1v4C", "object": "chat.completion", "created": 1727346166, "model": "gpt-4o-2024-08-06", "choices": [{"index": 0, "message": {"role": "assistant", "content": null, "tool_calls": [{"id": "call_fdNz3vOBKYgOIpMdWotB9MjY", "type": "function", "function": {"name": "GetWeatherArgs", "arguments": "{\\"city\\": \\"Edinburgh\\", \\"country\\": \\"GB\\", \\"units\\": \\"c\\"}"}}, {"id": "call_h1DWI1POMJLb0KwIyQHWXD4p", "type": "function", "function": {"name": "get_stock_price", "arguments": "{\\"ticker\\": \\"AAPL\\", \\"exchange\\": \\"NASDAQ\\"}"}}], "refusal": null}, "logprobs": null, "finish_reason": "tool_calls"}], "usage": {"prompt_tokens": 149, "completion_tokens": 60, "total_tokens": 209, "completion_tokens_details": {"reasoning_tokens": 0}}, "system_fingerprint": "fp_b40fb1c6fb"}'
         ),
+        path="/chat/completions",
         mock_client=client,
         respx_mock=respx_mock,
     )
@@ -733,7 +742,7 @@ class GetStockPrice(BaseModel):
 
 @pytest.mark.respx(base_url=base_url)
 def test_parse_strict_tools(client: OpenAI, respx_mock: MockRouter, monkeypatch: pytest.MonkeyPatch) -> None:
-    completion = _make_snapshot_request(
+    completion = make_snapshot_request(
         lambda c: c.chat.completions.parse(
             model="gpt-4o-2024-08-06",
             messages=[
@@ -767,6 +776,7 @@ def test_parse_strict_tools(client: OpenAI, respx_mock: MockRouter, monkeypatch:
         content_snapshot=snapshot(
             '{"id": "chatcmpl-ABfvzdvCI6RaIkiEFNjqGXCSYnlzf", "object": "chat.completion", "created": 1727346167, "model": "gpt-4o-2024-08-06", "choices": [{"index": 0, "message": {"role": "assistant", "content": null, "tool_calls": [{"id": "call_CUdUoJpsWWVdxXntucvnol1M", "type": "function", "function": {"name": "get_weather", "arguments": "{\\"city\\":\\"San Francisco\\",\\"state\\":\\"CA\\"}"}}], "refusal": null}, "logprobs": null, "finish_reason": "tool_calls"}], "usage": {"prompt_tokens": 48, "completion_tokens": 19, "total_tokens": 67, "completion_tokens_details": {"reasoning_tokens": 0}}, "system_fingerprint": "fp_5050236cbd"}'
         ),
+        path="/chat/completions",
         mock_client=client,
         respx_mock=respx_mock,
     )
@@ -830,7 +840,7 @@ class Location(BaseModel):
         temperature: float
         units: Literal["c", "f"]
 
-    response = _make_snapshot_request(
+    response = make_snapshot_request(
         lambda c: c.chat.completions.with_raw_response.parse(
             model="gpt-4o-2024-08-06",
             messages=[
@@ -844,6 +854,7 @@ class Location(BaseModel):
         content_snapshot=snapshot(
             '{"id": "chatcmpl-ABrDYCa8W1w66eUxKDO8TQF1m6trT", "object": "chat.completion", "created": 1727389540, "model": "gpt-4o-2024-08-06", "choices": [{"index": 0, "message": {"role": "assistant", "content": "{\\"city\\":\\"San Francisco\\",\\"temperature\\":58,\\"units\\":\\"f\\"}", "refusal": null}, "logprobs": null, "finish_reason": "stop"}], "usage": {"prompt_tokens": 79, "completion_tokens": 14, "total_tokens": 93, "completion_tokens_details": {"reasoning_tokens": 0}}, "system_fingerprint": "fp_5050236cbd"}'
         ),
+        path="/chat/completions",
         mock_client=client,
         respx_mock=respx_mock,
     )
@@ -906,7 +917,7 @@ class Location(BaseModel):
         temperature: float
         units: Literal["c", "f"]
 
-    response = await _make_async_snapshot_request(
+    response = await make_async_snapshot_request(
         lambda c: c.chat.completions.with_raw_response.parse(
             model="gpt-4o-2024-08-06",
             messages=[
@@ -920,6 +931,7 @@ class Location(BaseModel):
         content_snapshot=snapshot(
             '{"id": "chatcmpl-ABrDQWOiw0PK5JOsxl1D9ooeQgznq", "object": "chat.completion", "created": 1727389532, "model": "gpt-4o-2024-08-06", "choices": [{"index": 0, "message": {"role": "assistant", "content": "{\\"city\\":\\"San Francisco\\",\\"temperature\\":65,\\"units\\":\\"f\\"}", "refusal": null}, "logprobs": null, "finish_reason": "stop"}], "usage": {"prompt_tokens": 79, "completion_tokens": 14, "total_tokens": 93, "completion_tokens_details": {"reasoning_tokens": 0}}, "system_fingerprint": "fp_5050236cbd"}'
         ),
+        path="/chat/completions",
         mock_client=async_client,
         respx_mock=respx_mock,
     )
@@ -981,87 +993,3 @@ def test_parse_method_in_sync(sync: bool, client: OpenAI, async_client: AsyncOpe
         checking_client.chat.completions.parse,
         exclude_params={"response_format", "stream"},
     )
-
-
-def _make_snapshot_request(
-    func: Callable[[OpenAI], _T],
-    *,
-    content_snapshot: Any,
-    respx_mock: MockRouter,
-    mock_client: OpenAI,
-) -> _T:
-    live = os.environ.get("OPENAI_LIVE") == "1"
-    if live:
-
-        def _on_response(response: httpx.Response) -> None:
-            # update the content snapshot
-            assert json.dumps(json.loads(response.read())) == content_snapshot
-
-        respx_mock.stop()
-
-        client = OpenAI(
-            http_client=httpx.Client(
-                event_hooks={
-                    "response": [_on_response],
-                }
-            )
-        )
-    else:
-        respx_mock.post("/chat/completions").mock(
-            return_value=httpx.Response(
-                200,
-                content=get_snapshot_value(content_snapshot),
-                headers={"content-type": "application/json"},
-            )
-        )
-
-        client = mock_client
-
-    result = func(client)
-
-    if live:
-        client.close()
-
-    return result
-
-
-async def _make_async_snapshot_request(
-    func: Callable[[AsyncOpenAI], Awaitable[_T]],
-    *,
-    content_snapshot: Any,
-    respx_mock: MockRouter,
-    mock_client: AsyncOpenAI,
-) -> _T:
-    live = os.environ.get("OPENAI_LIVE") == "1"
-    if live:
-
-        async def _on_response(response: httpx.Response) -> None:
-            # update the content snapshot
-            assert json.dumps(json.loads(await response.aread())) == content_snapshot
-
-        respx_mock.stop()
-
-        client = AsyncOpenAI(
-            http_client=httpx.AsyncClient(
-                event_hooks={
-                    "response": [_on_response],
-                }
-            )
-        )
-    else:
-        respx_mock.post("/chat/completions").mock(
-            return_value=httpx.Response(
-                200,
-                content=get_snapshot_value(content_snapshot),
-                headers={"content-type": "application/json"},
-            )
-        )
-
-        client = mock_client
-
-    result = await func(client)
-
-    if live:
-        await client.close()
-
-    return result
diff --git a/tests/lib/chat/test_completions_streaming.py b/tests/lib/chat/test_completions_streaming.py
index 1daa98c6a0..fa17f67177 100644
--- a/tests/lib/chat/test_completions_streaming.py
+++ b/tests/lib/chat/test_completions_streaming.py
@@ -30,7 +30,7 @@
 )
 from openai.lib._parsing._completions import ResponseFormatT
 
-from ._utils import print_obj, get_snapshot_value
+from ..utils import print_obj, get_snapshot_value
 from ...conftest import base_url
 
 _T = TypeVar("_T")
@@ -39,7 +39,7 @@
 #
 # you can update them with
 #
-# `OPENAI_LIVE=1 pytest --inline-snapshot=fix`
+# `OPENAI_LIVE=1 pytest --inline-snapshot=fix -p no:xdist -o addopts=""`
 
 
 @pytest.mark.respx(base_url=base_url)
diff --git a/tests/lib/responses/__init__.py b/tests/lib/responses/__init__.py
new file mode 100644
index 0000000000..e69de29bb2
diff --git a/tests/lib/responses/test_responses.py b/tests/lib/responses/test_responses.py
new file mode 100644
index 0000000000..8ce3462e76
--- /dev/null
+++ b/tests/lib/responses/test_responses.py
@@ -0,0 +1,40 @@
+from __future__ import annotations
+
+from typing_extensions import TypeVar
+
+import pytest
+from respx import MockRouter
+from inline_snapshot import snapshot
+
+from openai import OpenAI
+
+from ...conftest import base_url
+from ..snapshots import make_snapshot_request
+
+_T = TypeVar("_T")
+
+# all the snapshots in this file are auto-generated from the live API
+#
+# you can update them with
+#
+# `OPENAI_LIVE=1 pytest --inline-snapshot=fix -p no:xdist -o addopts=""`
+
+
+@pytest.mark.respx(base_url=base_url)
+def test_output_text(client: OpenAI, respx_mock: MockRouter) -> None:
+    response = make_snapshot_request(
+        lambda c: c.responses.create(
+            model="gpt-4o-mini",
+            input="What's the weather like in SF?",
+        ),
+        content_snapshot=snapshot(
+            '{"id": "resp_689a0b2545288193953c892439b42e2800b2e36c65a1fd4b", "object": "response", "created_at": 1754925861, "status": "completed", "background": false, "error": null, "incomplete_details": null, "instructions": null, "max_output_tokens": null, "max_tool_calls": null, "model": "gpt-4o-mini-2024-07-18", "output": [{"id": "msg_689a0b2637b08193ac478e568f49e3f900b2e36c65a1fd4b", "type": "message", "status": "completed", "content": [{"type": "output_text", "annotations": [], "logprobs": [], "text": "I can\'t provide real-time updates, but you can easily check the current weather in San Francisco using a weather website or app. Typically, San Francisco has cool, foggy summers and mild winters, so it\'s good to be prepared for variable weather!"}], "role": "assistant"}], "parallel_tool_calls": true, "previous_response_id": null, "prompt_cache_key": null, "reasoning": {"effort": null, "summary": null}, "safety_identifier": null, "service_tier": "default", "store": true, "temperature": 1.0, "text": {"format": {"type": "text"}, "verbosity": "medium"}, "tool_choice": "auto", "tools": [], "top_logprobs": 0, "top_p": 1.0, "truncation": "disabled", "usage": {"input_tokens": 14, "input_tokens_details": {"cached_tokens": 0}, "output_tokens": 50, "output_tokens_details": {"reasoning_tokens": 0}, "total_tokens": 64}, "user": null, "metadata": {}}'
+        ),
+        path="/responses",
+        mock_client=client,
+        respx_mock=respx_mock,
+    )
+
+    assert response.output_text == snapshot(
+        "I can't provide real-time updates, but you can easily check the current weather in San Francisco using a weather website or app. Typically, San Francisco has cool, foggy summers and mild winters, so it's good to be prepared for variable weather!"
+    )
diff --git a/tests/lib/snapshots.py b/tests/lib/snapshots.py
new file mode 100644
index 0000000000..ed53edebcb
--- /dev/null
+++ b/tests/lib/snapshots.py
@@ -0,0 +1,101 @@
+from __future__ import annotations
+
+import os
+import json
+from typing import Any, Callable, Awaitable
+from typing_extensions import TypeVar
+
+import httpx
+from respx import MockRouter
+
+from openai import OpenAI, AsyncOpenAI
+
+from .utils import get_snapshot_value
+
+_T = TypeVar("_T")
+
+
+def make_snapshot_request(
+    func: Callable[[OpenAI], _T],
+    *,
+    content_snapshot: Any,
+    respx_mock: MockRouter,
+    mock_client: OpenAI,
+    path: str,
+) -> _T:
+    live = os.environ.get("OPENAI_LIVE") == "1"
+    if live:
+
+        def _on_response(response: httpx.Response) -> None:
+            # update the content snapshot
+            assert json.dumps(json.loads(response.read())) == content_snapshot
+
+        respx_mock.stop()
+
+        client = OpenAI(
+            http_client=httpx.Client(
+                event_hooks={
+                    "response": [_on_response],
+                }
+            )
+        )
+    else:
+        respx_mock.post(path).mock(
+            return_value=httpx.Response(
+                200,
+                content=get_snapshot_value(content_snapshot),
+                headers={"content-type": "application/json"},
+            )
+        )
+
+        client = mock_client
+
+    result = func(client)
+
+    if live:
+        client.close()
+
+    return result
+
+
+async def make_async_snapshot_request(
+    func: Callable[[AsyncOpenAI], Awaitable[_T]],
+    *,
+    content_snapshot: Any,
+    respx_mock: MockRouter,
+    mock_client: AsyncOpenAI,
+    path: str,
+) -> _T:
+    live = os.environ.get("OPENAI_LIVE") == "1"
+    if live:
+
+        async def _on_response(response: httpx.Response) -> None:
+            # update the content snapshot
+            assert json.dumps(json.loads(await response.aread())) == content_snapshot
+
+        respx_mock.stop()
+
+        client = AsyncOpenAI(
+            http_client=httpx.AsyncClient(
+                event_hooks={
+                    "response": [_on_response],
+                }
+            )
+        )
+    else:
+        respx_mock.post(path).mock(
+            return_value=httpx.Response(
+                200,
+                content=get_snapshot_value(content_snapshot),
+                headers={"content-type": "application/json"},
+            )
+        )
+
+        client = mock_client
+
+    result = await func(client)
+
+    if live:
+        await client.close()
+
+    return result
diff --git a/tests/lib/chat/_utils.py b/tests/lib/utils.py
similarity index 98%
rename from tests/lib/chat/_utils.py
rename to tests/lib/utils.py
index 0cc1c99952..2129ee811a 100644
--- a/tests/lib/chat/_utils.py
+++ b/tests/lib/utils.py
@@ -7,7 +7,7 @@
 import pytest
 import pydantic
 
-from ...utils import rich_print_str
+from ..utils import rich_print_str
 
 ReprArgs: TypeAlias = "Iterable[tuple[str | None, Any]]"