diff --git a/.release-please-manifest.json b/.release-please-manifest.json
index 61d831bd..f04d0896 100644
--- a/.release-please-manifest.json
+++ b/.release-please-manifest.json
@@ -1,3 +1,3 @@
 {
-  ".": "0.31.2"
+  ".": "0.32.0"
 }
\ No newline at end of file
diff --git a/.stats.yml b/.stats.yml
index 3ad8fd53..ff2805b6 100644
--- a/.stats.yml
+++ b/.stats.yml
@@ -1,2 +1,2 @@
 configured_endpoints: 2
-openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/anthropic-e2a51f04a202c13736b6fa2061a89a0c443f99ab166d965d702baf371eb1ca8f.yml
+openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/anthropic-5903ec2fd4efd7f261908bc4ec8ecd6b19cb9efa79637ad273583f1b763f80fd.yml
diff --git a/CHANGELOG.md b/CHANGELOG.md
index eb6ddbf4..516cd35c 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,5 +1,32 @@
 # Changelog
 
+## 0.32.0 (2024-07-29)
+
+Full Changelog: [v0.31.2...v0.32.0](https://github.com/anthropics/anthropic-sdk-python/compare/v0.31.2...v0.32.0)
+
+### Features
+
+* add back compat alias for InputJsonDelta ([25a5b6c](https://github.com/anthropics/anthropic-sdk-python/commit/25a5b6c81ffb5996ef697aab22a22d8be5751bc1))
+
+
+### Bug Fixes
+
+* change signatures for the stream function ([c9eb11b](https://github.com/anthropics/anthropic-sdk-python/commit/c9eb11b1f9656202ee88e9869e59160bc37f5434))
+* **client:** correctly apply client level timeout for messages ([#615](https://github.com/anthropics/anthropic-sdk-python/issues/615)) ([5f8d88f](https://github.com/anthropics/anthropic-sdk-python/commit/5f8d88f6fcc2ba05cd9fc6f8ae7aa8c61dc6b0d0))
+
+
+### Chores
+
+* **docs:** document how to do per-request http client customization ([#603](https://github.com/anthropics/anthropic-sdk-python/issues/603)) ([5161f62](https://github.com/anthropics/anthropic-sdk-python/commit/5161f626a0bec757b96217dc0f81e8908546f29a))
+* **internal:** add type construction helper ([#613](https://github.com/anthropics/anthropic-sdk-python/issues/613)) ([5e36940](https://github.com/anthropics/anthropic-sdk-python/commit/5e36940a42e401c3f0c1e42aa248d431fdf7192c))
+* sync spec ([#605](https://github.com/anthropics/anthropic-sdk-python/issues/605)) ([6b7707f](https://github.com/anthropics/anthropic-sdk-python/commit/6b7707f62788fca2e166209e82935a2a2fa8204a))
+* **tests:** update prism version ([#607](https://github.com/anthropics/anthropic-sdk-python/issues/607)) ([1797dc6](https://github.com/anthropics/anthropic-sdk-python/commit/1797dc6139ffaca6436ed897972471e67ba1b828))
+
+
+### Refactors
+
+* extract model out to a named type and rename partialjson ([#612](https://github.com/anthropics/anthropic-sdk-python/issues/612)) ([c53efc7](https://github.com/anthropics/anthropic-sdk-python/commit/c53efc786fa95831a398f37740a81b42f7b64c94))
+
 ## 0.31.2 (2024-07-17)
 
 Full Changelog: [v0.31.1...v0.31.2](https://github.com/anthropics/anthropic-sdk-python/compare/v0.31.1...v0.31.2)
diff --git a/README.md b/README.md
index fcacca0f..0fd0f870 100644
--- a/README.md
+++ b/README.md
@@ -508,6 +508,12 @@ client = Anthropic(
 )
 ```
 
+You can also customize the client on a per-request basis by using `with_options()`:
+
+```python
+client.with_options(http_client=DefaultHttpxClient(...))
+```
+
 ### Managing HTTP resources
 
 By default the library closes underlying HTTP connections whenever the client is [garbage collected](https://docs.python.org/3/reference/datamodel.html#object.__del__). You can manually close the client using the `.close()` method if desired, or with a context manager that closes when exiting.
diff --git a/api.md b/api.md
index 5ac54252..63896541 100644
--- a/api.md
+++ b/api.md
@@ -9,7 +9,7 @@ from anthropic.types import (
     ContentBlockStartEvent,
     ContentBlockStopEvent,
     ImageBlockParam,
-    InputJsonDelta,
+    InputJSONDelta,
     Message,
     MessageDeltaEvent,
     MessageDeltaUsage,
@@ -17,6 +17,7 @@ from anthropic.types import (
     MessageStartEvent,
     MessageStopEvent,
     MessageStreamEvent,
+    Model,
     RawContentBlockDeltaEvent,
     RawContentBlockStartEvent,
     RawContentBlockStopEvent,
diff --git a/pyproject.toml b/pyproject.toml
index 10921bf7..ac92a41f 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -1,6 +1,6 @@
 [project]
 name = "anthropic"
-version = "0.31.2"
+version = "0.32.0"
 description = "The official Python library for the anthropic API"
 dynamic = ["readme"]
 license = "MIT"
diff --git a/scripts/mock b/scripts/mock
index fe89a1d0..f5861576 100755
--- a/scripts/mock
+++ b/scripts/mock
@@ -21,7 +21,7 @@ echo "==> Starting mock server with URL ${URL}"
 
 # Run prism mock on the given spec
 if [ "$1" == "--daemon" ]; then
-  npm exec --package=@stoplight/prism-cli@~5.8 -- prism mock "$URL" &> .prism.log &
+  npm exec --package=@stainless-api/prism-cli@5.8.4 -- prism mock "$URL" &> .prism.log &
 
   # Wait for server to come online
   echo -n "Waiting for server"
@@ -37,5 +37,5 @@ if [ "$1" == "--daemon" ]; then
 
   echo
 else
-  npm exec  --package=@stoplight/prism-cli@~5.8 -- prism mock "$URL"
+  npm exec --package=@stainless-api/prism-cli@5.8.4 -- prism mock "$URL"
 fi
diff --git a/src/anthropic/_models.py b/src/anthropic/_models.py
index eb7ce3bd..5148d5a7 100644
--- a/src/anthropic/_models.py
+++ b/src/anthropic/_models.py
@@ -406,6 +406,15 @@ def build(
     return cast(_BaseModelT, construct_type(type_=base_model_cls, value=kwargs))
 
 
+def construct_type_unchecked(*, value: object, type_: type[_T]) -> _T:
+    """Loose coercion to the expected type with construction of nested values.
+
+    Note: the returned value from this function is not guaranteed to match the
+    given type.
+    """
+    return cast(_T, construct_type(value=value, type_=type_))
+
+
 def construct_type(*, value: object, type_: object) -> object:
     """Loose coercion to the expected type with construction of nested values.
 
diff --git a/src/anthropic/_utils/_reflection.py b/src/anthropic/_utils/_reflection.py
index 9a53c7bd..89aa712a 100644
--- a/src/anthropic/_utils/_reflection.py
+++ b/src/anthropic/_utils/_reflection.py
@@ -34,7 +34,7 @@ def assert_signatures_in_sync(
 
         if custom_param.annotation != source_param.annotation:
             errors.append(
-                f"types for the `{name}` param are do not match; source={repr(source_param.annotation)} checking={repr(source_param.annotation)}"
+                f"types for the `{name}` param are do not match; source={repr(source_param.annotation)} checking={repr(custom_param.annotation)}"
             )
             continue
 
diff --git a/src/anthropic/_version.py b/src/anthropic/_version.py
index 040dba2e..722453a0 100644
--- a/src/anthropic/_version.py
+++ b/src/anthropic/_version.py
@@ -1,4 +1,4 @@
 # File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
 
 __title__ = "anthropic"
-__version__ = "0.31.2"  # x-release-please-version
+__version__ = "0.32.0"  # x-release-please-version
diff --git a/src/anthropic/resources/completions.py b/src/anthropic/resources/completions.py
index f0c9afaf..7369f485 100644
--- a/src/anthropic/resources/completions.py
+++ b/src/anthropic/resources/completions.py
@@ -2,7 +2,7 @@
 
 from __future__ import annotations
 
-from typing import List, Union, overload
+from typing import List, overload
 from typing_extensions import Literal
 
 import httpx
@@ -11,6 +11,7 @@
 from ..types import completion_create_params
 from .._types import NOT_GIVEN, Body, Query, Headers, NotGiven
 from .._utils import (
+    is_given,
     required_args,
     maybe_transform,
     async_maybe_transform,
@@ -18,9 +19,11 @@
 from .._compat import cached_property
 from .._resource import SyncAPIResource, AsyncAPIResource
 from .._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
+from .._constants import DEFAULT_TIMEOUT
 from .._streaming import Stream, AsyncStream
 from .._base_client import make_request_options
 from ..types.completion import Completion
+from ..types.model_param import ModelParam
 
 __all__ = ["Completions", "AsyncCompletions"]
 
@@ -39,7 +42,7 @@ def create(
         self,
         *,
         max_tokens_to_sample: int,
-        model: Union[str, Literal["claude-2.0", "claude-2.1", "claude-instant-1.2"]],
+        model: ModelParam,
         prompt: str,
         metadata: completion_create_params.Metadata | NotGiven = NOT_GIVEN,
         stop_sequences: List[str] | NotGiven = NOT_GIVEN,
@@ -52,7 +55,7 @@ def create(
         extra_headers: Headers | None = None,
         extra_query: Query | None = None,
         extra_body: Body | None = None,
-        timeout: float | httpx.Timeout | None | NotGiven = 600,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> Completion:
         """[Legacy] Create a Text Completion.
 
@@ -71,9 +74,8 @@ def create(
               Note that our models may stop _before_ reaching this maximum. This parameter
               only specifies the absolute maximum number of tokens to generate.
 
-          model: The model that will complete your prompt.
-
-              See [models](https://docs.anthropic.com/en/docs/models-overview) for additional
+          model: The model that will complete your prompt.\n\nSee
+              [models](https://docs.anthropic.com/en/docs/models-overview) for additional
               details and options.
 
           prompt: The prompt that you want Claude to complete.
@@ -144,7 +146,7 @@ def create(
         self,
         *,
         max_tokens_to_sample: int,
-        model: Union[str, Literal["claude-2.0", "claude-2.1", "claude-instant-1.2"]],
+        model: ModelParam,
         prompt: str,
         stream: Literal[True],
         metadata: completion_create_params.Metadata | NotGiven = NOT_GIVEN,
@@ -157,7 +159,7 @@ def create(
         extra_headers: Headers | None = None,
         extra_query: Query | None = None,
         extra_body: Body | None = None,
-        timeout: float | httpx.Timeout | None | NotGiven = 600,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> Stream[Completion]:
         """[Legacy] Create a Text Completion.
 
@@ -176,9 +178,8 @@ def create(
               Note that our models may stop _before_ reaching this maximum. This parameter
               only specifies the absolute maximum number of tokens to generate.
 
-          model: The model that will complete your prompt.
-
-              See [models](https://docs.anthropic.com/en/docs/models-overview) for additional
+          model: The model that will complete your prompt.\n\nSee
+              [models](https://docs.anthropic.com/en/docs/models-overview) for additional
               details and options.
 
           prompt: The prompt that you want Claude to complete.
@@ -249,7 +250,7 @@ def create(
         self,
         *,
         max_tokens_to_sample: int,
-        model: Union[str, Literal["claude-2.0", "claude-2.1", "claude-instant-1.2"]],
+        model: ModelParam,
         prompt: str,
         stream: bool,
         metadata: completion_create_params.Metadata | NotGiven = NOT_GIVEN,
@@ -262,7 +263,7 @@ def create(
         extra_headers: Headers | None = None,
         extra_query: Query | None = None,
         extra_body: Body | None = None,
-        timeout: float | httpx.Timeout | None | NotGiven = 600,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> Completion | Stream[Completion]:
         """[Legacy] Create a Text Completion.
 
@@ -281,9 +282,8 @@ def create(
               Note that our models may stop _before_ reaching this maximum. This parameter
               only specifies the absolute maximum number of tokens to generate.
 
-          model: The model that will complete your prompt.
-
-              See [models](https://docs.anthropic.com/en/docs/models-overview) for additional
+          model: The model that will complete your prompt.\n\nSee
+              [models](https://docs.anthropic.com/en/docs/models-overview) for additional
               details and options.
 
           prompt: The prompt that you want Claude to complete.
@@ -354,7 +354,7 @@ def create(
         self,
         *,
         max_tokens_to_sample: int,
-        model: Union[str, Literal["claude-2.0", "claude-2.1", "claude-instant-1.2"]],
+        model: ModelParam,
         prompt: str,
         metadata: completion_create_params.Metadata | NotGiven = NOT_GIVEN,
         stop_sequences: List[str] | NotGiven = NOT_GIVEN,
@@ -367,8 +367,10 @@ def create(
         extra_headers: Headers | None = None,
         extra_query: Query | None = None,
         extra_body: Body | None = None,
-        timeout: float | httpx.Timeout | None | NotGiven = 600,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> Completion | Stream[Completion]:
+        if not is_given(timeout) and self._client.timeout == DEFAULT_TIMEOUT:
+            timeout = 600
         return self._post(
             "/v1/complete",
             body=maybe_transform(
@@ -408,7 +410,7 @@ async def create(
         self,
         *,
         max_tokens_to_sample: int,
-        model: Union[str, Literal["claude-2.0", "claude-2.1", "claude-instant-1.2"]],
+        model: ModelParam,
         prompt: str,
         metadata: completion_create_params.Metadata | NotGiven = NOT_GIVEN,
         stop_sequences: List[str] | NotGiven = NOT_GIVEN,
@@ -421,7 +423,7 @@ async def create(
         extra_headers: Headers | None = None,
         extra_query: Query | None = None,
         extra_body: Body | None = None,
-        timeout: float | httpx.Timeout | None | NotGiven = 600,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> Completion:
         """[Legacy] Create a Text Completion.
 
@@ -440,9 +442,8 @@ async def create(
               Note that our models may stop _before_ reaching this maximum. This parameter
               only specifies the absolute maximum number of tokens to generate.
 
-          model: The model that will complete your prompt.
-
-              See [models](https://docs.anthropic.com/en/docs/models-overview) for additional
+          model: The model that will complete your prompt.\n\nSee
+              [models](https://docs.anthropic.com/en/docs/models-overview) for additional
               details and options.
 
           prompt: The prompt that you want Claude to complete.
@@ -513,7 +514,7 @@ async def create(
         self,
         *,
         max_tokens_to_sample: int,
-        model: Union[str, Literal["claude-2.0", "claude-2.1", "claude-instant-1.2"]],
+        model: ModelParam,
         prompt: str,
         stream: Literal[True],
         metadata: completion_create_params.Metadata | NotGiven = NOT_GIVEN,
@@ -526,7 +527,7 @@ async def create(
         extra_headers: Headers | None = None,
         extra_query: Query | None = None,
         extra_body: Body | None = None,
-        timeout: float | httpx.Timeout | None | NotGiven = 600,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> AsyncStream[Completion]:
         """[Legacy] Create a Text Completion.
 
@@ -545,9 +546,8 @@ async def create(
               Note that our models may stop _before_ reaching this maximum. This parameter
               only specifies the absolute maximum number of tokens to generate.
 
-          model: The model that will complete your prompt.
-
-              See [models](https://docs.anthropic.com/en/docs/models-overview) for additional
+          model: The model that will complete your prompt.\n\nSee
+              [models](https://docs.anthropic.com/en/docs/models-overview) for additional
               details and options.
 
           prompt: The prompt that you want Claude to complete.
@@ -618,7 +618,7 @@ async def create(
         self,
         *,
         max_tokens_to_sample: int,
-        model: Union[str, Literal["claude-2.0", "claude-2.1", "claude-instant-1.2"]],
+        model: ModelParam,
         prompt: str,
         stream: bool,
         metadata: completion_create_params.Metadata | NotGiven = NOT_GIVEN,
@@ -631,7 +631,7 @@ async def create(
         extra_headers: Headers | None = None,
         extra_query: Query | None = None,
         extra_body: Body | None = None,
-        timeout: float | httpx.Timeout | None | NotGiven = 600,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> Completion | AsyncStream[Completion]:
         """[Legacy] Create a Text Completion.
 
@@ -650,9 +650,8 @@ async def create(
               Note that our models may stop _before_ reaching this maximum. This parameter
               only specifies the absolute maximum number of tokens to generate.
 
-          model: The model that will complete your prompt.
-
-              See [models](https://docs.anthropic.com/en/docs/models-overview) for additional
+          model: The model that will complete your prompt.\n\nSee
+              [models](https://docs.anthropic.com/en/docs/models-overview) for additional
               details and options.
 
           prompt: The prompt that you want Claude to complete.
@@ -723,7 +722,7 @@ async def create(
         self,
         *,
         max_tokens_to_sample: int,
-        model: Union[str, Literal["claude-2.0", "claude-2.1", "claude-instant-1.2"]],
+        model: ModelParam,
         prompt: str,
         metadata: completion_create_params.Metadata | NotGiven = NOT_GIVEN,
         stop_sequences: List[str] | NotGiven = NOT_GIVEN,
@@ -736,8 +735,10 @@ async def create(
         extra_headers: Headers | None = None,
         extra_query: Query | None = None,
         extra_body: Body | None = None,
-        timeout: float | httpx.Timeout | None | NotGiven = 600,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> Completion | AsyncStream[Completion]:
+        if not is_given(timeout) and self._client.timeout == DEFAULT_TIMEOUT:
+            timeout = 600
         return await self._post(
             "/v1/complete",
             body=await async_maybe_transform(
diff --git a/src/anthropic/resources/messages.py b/src/anthropic/resources/messages.py
index c7d3639f..8b6ab106 100644
--- a/src/anthropic/resources/messages.py
+++ b/src/anthropic/resources/messages.py
@@ -12,6 +12,7 @@
 from ..types import message_create_params
 from .._types import NOT_GIVEN, Body, Query, Headers, NotGiven
 from .._utils import (
+    is_given,
     required_args,
     maybe_transform,
     async_maybe_transform,
@@ -19,11 +20,13 @@
 from .._compat import cached_property
 from .._resource import SyncAPIResource, AsyncAPIResource
 from .._response import to_streamed_response_wrapper, async_to_streamed_response_wrapper
+from .._constants import DEFAULT_TIMEOUT
 from .._streaming import Stream, AsyncStream
 from .._base_client import make_request_options
 from ..lib.streaming import MessageStreamManager, AsyncMessageStreamManager
 from ..types.message import Message
 from ..types.tool_param import ToolParam
+from ..types.model_param import ModelParam
 from ..types.message_param import MessageParam
 from ..types.text_block_param import TextBlockParam
 from ..types.raw_message_stream_event import RawMessageStreamEvent
@@ -46,18 +49,7 @@ def create(
         *,
         max_tokens: int,
         messages: Iterable[MessageParam],
-        model: Union[
-            str,
-            Literal[
-                "claude-3-5-sonnet-20240620",
-                "claude-3-opus-20240229",
-                "claude-3-sonnet-20240229",
-                "claude-3-haiku-20240307",
-                "claude-2.1",
-                "claude-2.0",
-                "claude-instant-1.2",
-            ],
-        ],
+        model: ModelParam,
         metadata: message_create_params.Metadata | NotGiven = NOT_GIVEN,
         stop_sequences: List[str] | NotGiven = NOT_GIVEN,
         stream: Literal[False] | NotGiven = NOT_GIVEN,
@@ -72,7 +64,7 @@ def create(
         extra_headers: Headers | None = None,
         extra_query: Query | None = None,
         extra_body: Body | None = None,
-        timeout: float | httpx.Timeout | None | NotGiven = 600,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> Message:
         """
         Create a Message.
@@ -178,9 +170,8 @@ def create(
               the top-level `system` parameter — there is no `"system"` role for input
               messages in the Messages API.
 
-          model: The model that will complete your prompt.
-
-              See [models](https://docs.anthropic.com/en/docs/models-overview) for additional
+          model: The model that will complete your prompt.\n\nSee
+              [models](https://docs.anthropic.com/en/docs/models-overview) for additional
               details and options.
 
           metadata: An object describing metadata about the request.
@@ -321,18 +312,7 @@ def create(
         *,
         max_tokens: int,
         messages: Iterable[MessageParam],
-        model: Union[
-            str,
-            Literal[
-                "claude-3-5-sonnet-20240620",
-                "claude-3-opus-20240229",
-                "claude-3-sonnet-20240229",
-                "claude-3-haiku-20240307",
-                "claude-2.1",
-                "claude-2.0",
-                "claude-instant-1.2",
-            ],
-        ],
+        model: ModelParam,
         stream: Literal[True],
         metadata: message_create_params.Metadata | NotGiven = NOT_GIVEN,
         stop_sequences: List[str] | NotGiven = NOT_GIVEN,
@@ -347,7 +327,7 @@ def create(
         extra_headers: Headers | None = None,
         extra_query: Query | None = None,
         extra_body: Body | None = None,
-        timeout: float | httpx.Timeout | None | NotGiven = 600,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> Stream[RawMessageStreamEvent]:
         """
         Create a Message.
@@ -453,9 +433,8 @@ def create(
               the top-level `system` parameter — there is no `"system"` role for input
               messages in the Messages API.
 
-          model: The model that will complete your prompt.
-
-              See [models](https://docs.anthropic.com/en/docs/models-overview) for additional
+          model: The model that will complete your prompt.\n\nSee
+              [models](https://docs.anthropic.com/en/docs/models-overview) for additional
               details and options.
 
           stream: Whether to incrementally stream the response using server-sent events.
@@ -596,18 +575,7 @@ def create(
         *,
         max_tokens: int,
         messages: Iterable[MessageParam],
-        model: Union[
-            str,
-            Literal[
-                "claude-3-5-sonnet-20240620",
-                "claude-3-opus-20240229",
-                "claude-3-sonnet-20240229",
-                "claude-3-haiku-20240307",
-                "claude-2.1",
-                "claude-2.0",
-                "claude-instant-1.2",
-            ],
-        ],
+        model: ModelParam,
         stream: bool,
         metadata: message_create_params.Metadata | NotGiven = NOT_GIVEN,
         stop_sequences: List[str] | NotGiven = NOT_GIVEN,
@@ -622,7 +590,7 @@ def create(
         extra_headers: Headers | None = None,
         extra_query: Query | None = None,
         extra_body: Body | None = None,
-        timeout: float | httpx.Timeout | None | NotGiven = 600,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> Message | Stream[RawMessageStreamEvent]:
         """
         Create a Message.
@@ -728,9 +696,8 @@ def create(
               the top-level `system` parameter — there is no `"system"` role for input
               messages in the Messages API.
 
-          model: The model that will complete your prompt.
-
-              See [models](https://docs.anthropic.com/en/docs/models-overview) for additional
+          model: The model that will complete your prompt.\n\nSee
+              [models](https://docs.anthropic.com/en/docs/models-overview) for additional
               details and options.
 
           stream: Whether to incrementally stream the response using server-sent events.
@@ -871,18 +838,7 @@ def create(
         *,
         max_tokens: int,
         messages: Iterable[MessageParam],
-        model: Union[
-            str,
-            Literal[
-                "claude-3-5-sonnet-20240620",
-                "claude-3-opus-20240229",
-                "claude-3-sonnet-20240229",
-                "claude-3-haiku-20240307",
-                "claude-2.1",
-                "claude-2.0",
-                "claude-instant-1.2",
-            ],
-        ],
+        model: ModelParam,
         metadata: message_create_params.Metadata | NotGiven = NOT_GIVEN,
         stop_sequences: List[str] | NotGiven = NOT_GIVEN,
         stream: Literal[False] | Literal[True] | NotGiven = NOT_GIVEN,
@@ -897,8 +853,10 @@ def create(
         extra_headers: Headers | None = None,
         extra_query: Query | None = None,
         extra_body: Body | None = None,
-        timeout: float | httpx.Timeout | None | NotGiven = 600,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> Message | Stream[RawMessageStreamEvent]:
+        if not is_given(timeout) and self._client.timeout == DEFAULT_TIMEOUT:
+            timeout = 600
         return self._post(
             "/v1/messages",
             body=maybe_transform(
@@ -931,18 +889,7 @@ def stream(
         *,
         max_tokens: int,
         messages: Iterable[MessageParam],
-        model: Union[
-            str,
-            Literal[
-                "claude-3-5-sonnet-20240620",
-                "claude-3-opus-20240229",
-                "claude-3-sonnet-20240229",
-                "claude-3-haiku-20240307",
-                "claude-2.1",
-                "claude-2.0",
-                "claude-instant-1.2",
-            ],
-        ],
+        model: ModelParam,
         metadata: message_create_params.Metadata | NotGiven = NOT_GIVEN,
         stop_sequences: List[str] | NotGiven = NOT_GIVEN,
         system: Union[str, Iterable[TextBlockParam]] | NotGiven = NOT_GIVEN,
@@ -1008,18 +955,7 @@ async def create(
         *,
         max_tokens: int,
         messages: Iterable[MessageParam],
-        model: Union[
-            str,
-            Literal[
-                "claude-3-5-sonnet-20240620",
-                "claude-3-opus-20240229",
-                "claude-3-sonnet-20240229",
-                "claude-3-haiku-20240307",
-                "claude-2.1",
-                "claude-2.0",
-                "claude-instant-1.2",
-            ],
-        ],
+        model: ModelParam,
         metadata: message_create_params.Metadata | NotGiven = NOT_GIVEN,
         stop_sequences: List[str] | NotGiven = NOT_GIVEN,
         stream: Literal[False] | NotGiven = NOT_GIVEN,
@@ -1034,7 +970,7 @@ async def create(
         extra_headers: Headers | None = None,
         extra_query: Query | None = None,
         extra_body: Body | None = None,
-        timeout: float | httpx.Timeout | None | NotGiven = 600,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> Message:
         """
         Create a Message.
@@ -1140,9 +1076,8 @@ async def create(
               the top-level `system` parameter — there is no `"system"` role for input
               messages in the Messages API.
 
-          model: The model that will complete your prompt.
-
-              See [models](https://docs.anthropic.com/en/docs/models-overview) for additional
+          model: The model that will complete your prompt.\n\nSee
+              [models](https://docs.anthropic.com/en/docs/models-overview) for additional
               details and options.
 
           metadata: An object describing metadata about the request.
@@ -1283,18 +1218,7 @@ async def create(
         *,
         max_tokens: int,
         messages: Iterable[MessageParam],
-        model: Union[
-            str,
-            Literal[
-                "claude-3-5-sonnet-20240620",
-                "claude-3-opus-20240229",
-                "claude-3-sonnet-20240229",
-                "claude-3-haiku-20240307",
-                "claude-2.1",
-                "claude-2.0",
-                "claude-instant-1.2",
-            ],
-        ],
+        model: ModelParam,
         stream: Literal[True],
         metadata: message_create_params.Metadata | NotGiven = NOT_GIVEN,
         stop_sequences: List[str] | NotGiven = NOT_GIVEN,
@@ -1309,7 +1233,7 @@ async def create(
         extra_headers: Headers | None = None,
         extra_query: Query | None = None,
         extra_body: Body | None = None,
-        timeout: float | httpx.Timeout | None | NotGiven = 600,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> AsyncStream[RawMessageStreamEvent]:
         """
         Create a Message.
@@ -1415,9 +1339,8 @@ async def create(
               the top-level `system` parameter — there is no `"system"` role for input
               messages in the Messages API.
 
-          model: The model that will complete your prompt.
-
-              See [models](https://docs.anthropic.com/en/docs/models-overview) for additional
+          model: The model that will complete your prompt.\n\nSee
+              [models](https://docs.anthropic.com/en/docs/models-overview) for additional
               details and options.
 
           stream: Whether to incrementally stream the response using server-sent events.
@@ -1558,18 +1481,7 @@ async def create(
         *,
         max_tokens: int,
         messages: Iterable[MessageParam],
-        model: Union[
-            str,
-            Literal[
-                "claude-3-5-sonnet-20240620",
-                "claude-3-opus-20240229",
-                "claude-3-sonnet-20240229",
-                "claude-3-haiku-20240307",
-                "claude-2.1",
-                "claude-2.0",
-                "claude-instant-1.2",
-            ],
-        ],
+        model: ModelParam,
         stream: bool,
         metadata: message_create_params.Metadata | NotGiven = NOT_GIVEN,
         stop_sequences: List[str] | NotGiven = NOT_GIVEN,
@@ -1584,7 +1496,7 @@ async def create(
         extra_headers: Headers | None = None,
         extra_query: Query | None = None,
         extra_body: Body | None = None,
-        timeout: float | httpx.Timeout | None | NotGiven = 600,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> Message | AsyncStream[RawMessageStreamEvent]:
         """
         Create a Message.
@@ -1690,9 +1602,8 @@ async def create(
               the top-level `system` parameter — there is no `"system"` role for input
               messages in the Messages API.
 
-          model: The model that will complete your prompt.
-
-              See [models](https://docs.anthropic.com/en/docs/models-overview) for additional
+          model: The model that will complete your prompt.\n\nSee
+              [models](https://docs.anthropic.com/en/docs/models-overview) for additional
               details and options.
 
           stream: Whether to incrementally stream the response using server-sent events.
@@ -1833,18 +1744,7 @@ async def create(
         *,
         max_tokens: int,
         messages: Iterable[MessageParam],
-        model: Union[
-            str,
-            Literal[
-                "claude-3-5-sonnet-20240620",
-                "claude-3-opus-20240229",
-                "claude-3-sonnet-20240229",
-                "claude-3-haiku-20240307",
-                "claude-2.1",
-                "claude-2.0",
-                "claude-instant-1.2",
-            ],
-        ],
+        model: ModelParam,
         metadata: message_create_params.Metadata | NotGiven = NOT_GIVEN,
         stop_sequences: List[str] | NotGiven = NOT_GIVEN,
         stream: Literal[False] | Literal[True] | NotGiven = NOT_GIVEN,
@@ -1859,8 +1759,10 @@ async def create(
         extra_headers: Headers | None = None,
         extra_query: Query | None = None,
         extra_body: Body | None = None,
-        timeout: float | httpx.Timeout | None | NotGiven = 600,
+        timeout: float | httpx.Timeout | None | NotGiven = NOT_GIVEN,
     ) -> Message | AsyncStream[RawMessageStreamEvent]:
+        if not is_given(timeout) and self._client.timeout == DEFAULT_TIMEOUT:
+            timeout = 600
         return await self._post(
             "/v1/messages",
             body=await async_maybe_transform(
@@ -1893,18 +1795,7 @@ def stream(
         *,
         max_tokens: int,
         messages: Iterable[MessageParam],
-        model: Union[
-            str,
-            Literal[
-                "claude-3-5-sonnet-20240620",
-                "claude-3-opus-20240229",
-                "claude-3-sonnet-20240229",
-                "claude-3-haiku-20240307",
-                "claude-2.1",
-                "claude-2.0",
-                "claude-instant-1.2",
-            ],
-        ],
+        model: ModelParam,
         metadata: message_create_params.Metadata | NotGiven = NOT_GIVEN,
         stop_sequences: List[str] | NotGiven = NOT_GIVEN,
         system: Union[str, Iterable[TextBlockParam]] | NotGiven = NOT_GIVEN,
diff --git a/src/anthropic/types/__init__.py b/src/anthropic/types/__init__.py
index f6165eca..9839b5a2 100644
--- a/src/anthropic/types/__init__.py
+++ b/src/anthropic/types/__init__.py
@@ -2,16 +2,18 @@
 
 from __future__ import annotations
 
+from .model import Model as Model
 from .usage import Usage as Usage
 from .message import Message as Message
 from .completion import Completion as Completion
 from .text_block import TextBlock as TextBlock
 from .text_delta import TextDelta as TextDelta
 from .tool_param import ToolParam as ToolParam
+from .model_param import ModelParam as ModelParam
 from .content_block import ContentBlock as ContentBlock
 from .message_param import MessageParam as MessageParam
 from .tool_use_block import ToolUseBlock as ToolUseBlock
-from .input_json_delta import InputJsonDelta as InputJsonDelta
+from .input_json_delta import InputJSONDelta as InputJSONDelta
 from .text_block_param import TextBlockParam as TextBlockParam
 from .image_block_param import ImageBlockParam as ImageBlockParam
 from .message_stop_event import MessageStopEvent as MessageStopEvent
diff --git a/src/anthropic/types/completion.py b/src/anthropic/types/completion.py
index d55a0fe6..e6293210 100644
--- a/src/anthropic/types/completion.py
+++ b/src/anthropic/types/completion.py
@@ -3,6 +3,7 @@
 from typing import Optional
 from typing_extensions import Literal
 
+from .model import Model
 from .._models import BaseModel
 
 __all__ = ["Completion"]
@@ -18,8 +19,12 @@ class Completion(BaseModel):
     completion: str
     """The resulting completion up to and excluding the stop sequences."""
 
-    model: str
-    """The model that handled the request."""
+    model: Model
+    """
+    The model that will complete your prompt.\n\nSee
+    [models](https://docs.anthropic.com/en/docs/models-overview) for additional
+    details and options.
+    """
 
     stop_reason: Optional[str] = None
     """The reason that we stopped.
diff --git a/src/anthropic/types/completion_create_params.py b/src/anthropic/types/completion_create_params.py
index be824ebf..5c9ed7a6 100644
--- a/src/anthropic/types/completion_create_params.py
+++ b/src/anthropic/types/completion_create_params.py
@@ -5,6 +5,8 @@
 from typing import List, Union, Optional
 from typing_extensions import Literal, Required, TypedDict
 
+from .model_param import ModelParam
+
 __all__ = [
     "CompletionRequestStreamingMetadata",
     "CompletionRequestNonStreamingMetadata",
@@ -25,10 +27,10 @@ class CompletionCreateParamsBase(TypedDict, total=False):
     only specifies the absolute maximum number of tokens to generate.
     """
 
-    model: Required[Union[str, Literal["claude-2.0", "claude-2.1", "claude-instant-1.2"]]]
-    """The model that will complete your prompt.
-
-    See [models](https://docs.anthropic.com/en/docs/models-overview) for additional
+    model: Required[ModelParam]
+    """
+    The model that will complete your prompt.\n\nSee
+    [models](https://docs.anthropic.com/en/docs/models-overview) for additional
     details and options.
     """
 
diff --git a/src/anthropic/types/input_json_delta.py b/src/anthropic/types/input_json_delta.py
index 6391d4bf..3028d7ac 100644
--- a/src/anthropic/types/input_json_delta.py
+++ b/src/anthropic/types/input_json_delta.py
@@ -4,10 +4,12 @@
 
 from .._models import BaseModel
 
-__all__ = ["InputJsonDelta"]
+__all__ = ["InputJSONDelta", "InputJsonDelta"]
 
-
-class InputJsonDelta(BaseModel):
+class InputJSONDelta(BaseModel):
     partial_json: str
 
     type: Literal["input_json_delta"]
+
+InputJsonDelta = InputJSONDelta
+
diff --git a/src/anthropic/types/message.py b/src/anthropic/types/message.py
index 9ef967ea..3ddbf20b 100644
--- a/src/anthropic/types/message.py
+++ b/src/anthropic/types/message.py
@@ -3,6 +3,7 @@
 from typing import List, Optional
 from typing_extensions import Literal
 
+from .model import Model
 from .usage import Usage
 from .._models import BaseModel
 from .content_block import ContentBlock, ContentBlock as ContentBlock
@@ -52,8 +53,12 @@ class Message(BaseModel):
     ```
     """
 
-    model: str
-    """The model that handled the request."""
+    model: Model
+    """
+    The model that will complete your prompt.\n\nSee
+    [models](https://docs.anthropic.com/en/docs/models-overview) for additional
+    details and options.
+    """
 
     role: Literal["assistant"]
     """Conversational role of the generated message.
diff --git a/src/anthropic/types/message_create_params.py b/src/anthropic/types/message_create_params.py
index a76bc0f7..a50e5946 100644
--- a/src/anthropic/types/message_create_params.py
+++ b/src/anthropic/types/message_create_params.py
@@ -6,6 +6,7 @@
 from typing_extensions import Literal, Required, TypedDict
 
 from .tool_param import ToolParam
+from .model_param import ModelParam
 from .message_param import MessageParam
 from .text_block_param import TextBlockParam
 
@@ -120,23 +121,10 @@ class MessageCreateParamsBase(TypedDict, total=False):
     messages in the Messages API.
     """
 
-    model: Required[
-        Union[
-            str,
-            Literal[
-                "claude-3-5-sonnet-20240620",
-                "claude-3-opus-20240229",
-                "claude-3-sonnet-20240229",
-                "claude-3-haiku-20240307",
-                "claude-2.1",
-                "claude-2.0",
-                "claude-instant-1.2",
-            ],
-        ]
-    ]
-    """The model that will complete your prompt.
-
-    See [models](https://docs.anthropic.com/en/docs/models-overview) for additional
+    model: Required[ModelParam]
+    """
+    The model that will complete your prompt.\n\nSee
+    [models](https://docs.anthropic.com/en/docs/models-overview) for additional
     details and options.
     """
 
diff --git a/src/anthropic/types/model.py b/src/anthropic/types/model.py
new file mode 100644
index 00000000..0ada4aeb
--- /dev/null
+++ b/src/anthropic/types/model.py
@@ -0,0 +1,19 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from typing import Union
+from typing_extensions import Literal
+
+__all__ = ["Model"]
+
+Model = Union[
+    str,
+    Literal[
+        "claude-3-5-sonnet-20240620",
+        "claude-3-opus-20240229",
+        "claude-3-sonnet-20240229",
+        "claude-3-haiku-20240307",
+        "claude-2.1",
+        "claude-2.0",
+        "claude-instant-1.2",
+    ],
+]
diff --git a/src/anthropic/types/model_param.py b/src/anthropic/types/model_param.py
new file mode 100644
index 00000000..d932500f
--- /dev/null
+++ b/src/anthropic/types/model_param.py
@@ -0,0 +1,21 @@
+# File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.
+
+from __future__ import annotations
+
+from typing import Union
+from typing_extensions import Literal
+
+__all__ = ["ModelParam"]
+
+ModelParam = Union[
+    str,
+    Literal[
+        "claude-3-5-sonnet-20240620",
+        "claude-3-opus-20240229",
+        "claude-3-sonnet-20240229",
+        "claude-3-haiku-20240307",
+        "claude-2.1",
+        "claude-2.0",
+        "claude-instant-1.2",
+    ],
+]
diff --git a/src/anthropic/types/raw_content_block_delta_event.py b/src/anthropic/types/raw_content_block_delta_event.py
index e1370fdb..845d502a 100644
--- a/src/anthropic/types/raw_content_block_delta_event.py
+++ b/src/anthropic/types/raw_content_block_delta_event.py
@@ -6,11 +6,11 @@
 from .._utils import PropertyInfo
 from .._models import BaseModel
 from .text_delta import TextDelta
-from .input_json_delta import InputJsonDelta
+from .input_json_delta import InputJSONDelta
 
 __all__ = ["RawContentBlockDeltaEvent", "Delta"]
 
-Delta = Annotated[Union[TextDelta, InputJsonDelta], PropertyInfo(discriminator="type")]
+Delta = Annotated[Union[TextDelta, InputJSONDelta], PropertyInfo(discriminator="type")]
 
 
 class RawContentBlockDeltaEvent(BaseModel):
diff --git a/tests/api_resources/test_completions.py b/tests/api_resources/test_completions.py
index f46b7bb7..cf87d9fe 100644
--- a/tests/api_resources/test_completions.py
+++ b/tests/api_resources/test_completions.py
@@ -21,7 +21,7 @@ class TestCompletions:
     def test_method_create_overload_1(self, client: Anthropic) -> None:
         completion = client.completions.create(
             max_tokens_to_sample=256,
-            model="claude-2.1",
+            model="string",
             prompt="\n\nHuman: Hello, world!\n\nAssistant:",
         )
         assert_matches_type(Completion, completion, path=["response"])
@@ -30,7 +30,7 @@ def test_method_create_overload_1(self, client: Anthropic) -> None:
     def test_method_create_with_all_params_overload_1(self, client: Anthropic) -> None:
         completion = client.completions.create(
             max_tokens_to_sample=256,
-            model="claude-2.1",
+            model="string",
             prompt="\n\nHuman: Hello, world!\n\nAssistant:",
             metadata={"user_id": "13803d75-b4b5-4c3e-b2a2-6f21399b021b"},
             stop_sequences=["string", "string", "string"],
@@ -45,7 +45,7 @@ def test_method_create_with_all_params_overload_1(self, client: Anthropic) -> No
     def test_raw_response_create_overload_1(self, client: Anthropic) -> None:
         response = client.completions.with_raw_response.create(
             max_tokens_to_sample=256,
-            model="claude-2.1",
+            model="string",
             prompt="\n\nHuman: Hello, world!\n\nAssistant:",
         )
 
@@ -58,7 +58,7 @@ def test_raw_response_create_overload_1(self, client: Anthropic) -> None:
     def test_streaming_response_create_overload_1(self, client: Anthropic) -> None:
         with client.completions.with_streaming_response.create(
             max_tokens_to_sample=256,
-            model="claude-2.1",
+            model="string",
             prompt="\n\nHuman: Hello, world!\n\nAssistant:",
         ) as response:
             assert not response.is_closed
@@ -73,7 +73,7 @@ def test_streaming_response_create_overload_1(self, client: Anthropic) -> None:
     def test_method_create_overload_2(self, client: Anthropic) -> None:
         completion_stream = client.completions.create(
             max_tokens_to_sample=256,
-            model="claude-2.1",
+            model="string",
             prompt="\n\nHuman: Hello, world!\n\nAssistant:",
             stream=True,
         )
@@ -83,7 +83,7 @@ def test_method_create_overload_2(self, client: Anthropic) -> None:
     def test_method_create_with_all_params_overload_2(self, client: Anthropic) -> None:
         completion_stream = client.completions.create(
             max_tokens_to_sample=256,
-            model="claude-2.1",
+            model="string",
             prompt="\n\nHuman: Hello, world!\n\nAssistant:",
             stream=True,
             metadata={"user_id": "13803d75-b4b5-4c3e-b2a2-6f21399b021b"},
@@ -98,7 +98,7 @@ def test_method_create_with_all_params_overload_2(self, client: Anthropic) -> No
     def test_raw_response_create_overload_2(self, client: Anthropic) -> None:
         response = client.completions.with_raw_response.create(
             max_tokens_to_sample=256,
-            model="claude-2.1",
+            model="string",
             prompt="\n\nHuman: Hello, world!\n\nAssistant:",
             stream=True,
         )
@@ -111,7 +111,7 @@ def test_raw_response_create_overload_2(self, client: Anthropic) -> None:
     def test_streaming_response_create_overload_2(self, client: Anthropic) -> None:
         with client.completions.with_streaming_response.create(
             max_tokens_to_sample=256,
-            model="claude-2.1",
+            model="string",
             prompt="\n\nHuman: Hello, world!\n\nAssistant:",
             stream=True,
         ) as response:
@@ -131,7 +131,7 @@ class TestAsyncCompletions:
     async def test_method_create_overload_1(self, async_client: AsyncAnthropic) -> None:
         completion = await async_client.completions.create(
             max_tokens_to_sample=256,
-            model="claude-2.1",
+            model="string",
             prompt="\n\nHuman: Hello, world!\n\nAssistant:",
         )
         assert_matches_type(Completion, completion, path=["response"])
@@ -140,7 +140,7 @@ async def test_method_create_overload_1(self, async_client: AsyncAnthropic) -> N
     async def test_method_create_with_all_params_overload_1(self, async_client: AsyncAnthropic) -> None:
         completion = await async_client.completions.create(
             max_tokens_to_sample=256,
-            model="claude-2.1",
+            model="string",
             prompt="\n\nHuman: Hello, world!\n\nAssistant:",
             metadata={"user_id": "13803d75-b4b5-4c3e-b2a2-6f21399b021b"},
             stop_sequences=["string", "string", "string"],
@@ -155,7 +155,7 @@ async def test_method_create_with_all_params_overload_1(self, async_client: Asyn
     async def test_raw_response_create_overload_1(self, async_client: AsyncAnthropic) -> None:
         response = await async_client.completions.with_raw_response.create(
             max_tokens_to_sample=256,
-            model="claude-2.1",
+            model="string",
             prompt="\n\nHuman: Hello, world!\n\nAssistant:",
         )
 
@@ -168,7 +168,7 @@ async def test_raw_response_create_overload_1(self, async_client: AsyncAnthropic
     async def test_streaming_response_create_overload_1(self, async_client: AsyncAnthropic) -> None:
         async with async_client.completions.with_streaming_response.create(
             max_tokens_to_sample=256,
-            model="claude-2.1",
+            model="string",
             prompt="\n\nHuman: Hello, world!\n\nAssistant:",
         ) as response:
             assert not response.is_closed
@@ -183,7 +183,7 @@ async def test_streaming_response_create_overload_1(self, async_client: AsyncAnt
     async def test_method_create_overload_2(self, async_client: AsyncAnthropic) -> None:
         completion_stream = await async_client.completions.create(
             max_tokens_to_sample=256,
-            model="claude-2.1",
+            model="string",
             prompt="\n\nHuman: Hello, world!\n\nAssistant:",
             stream=True,
         )
@@ -193,7 +193,7 @@ async def test_method_create_overload_2(self, async_client: AsyncAnthropic) -> N
     async def test_method_create_with_all_params_overload_2(self, async_client: AsyncAnthropic) -> None:
         completion_stream = await async_client.completions.create(
             max_tokens_to_sample=256,
-            model="claude-2.1",
+            model="string",
             prompt="\n\nHuman: Hello, world!\n\nAssistant:",
             stream=True,
             metadata={"user_id": "13803d75-b4b5-4c3e-b2a2-6f21399b021b"},
@@ -208,7 +208,7 @@ async def test_method_create_with_all_params_overload_2(self, async_client: Asyn
     async def test_raw_response_create_overload_2(self, async_client: AsyncAnthropic) -> None:
         response = await async_client.completions.with_raw_response.create(
             max_tokens_to_sample=256,
-            model="claude-2.1",
+            model="string",
             prompt="\n\nHuman: Hello, world!\n\nAssistant:",
             stream=True,
         )
@@ -221,7 +221,7 @@ async def test_raw_response_create_overload_2(self, async_client: AsyncAnthropic
     async def test_streaming_response_create_overload_2(self, async_client: AsyncAnthropic) -> None:
         async with async_client.completions.with_streaming_response.create(
             max_tokens_to_sample=256,
-            model="claude-2.1",
+            model="string",
             prompt="\n\nHuman: Hello, world!\n\nAssistant:",
             stream=True,
         ) as response:
diff --git a/tests/lib/streaming/test_messages.py b/tests/lib/streaming/test_messages.py
index 408546de..e719516e 100644
--- a/tests/lib/streaming/test_messages.py
+++ b/tests/lib/streaming/test_messages.py
@@ -192,7 +192,7 @@ def test_stream_method_definition_in_sync(sync: bool) -> None:
 
         if custom_param.annotation != generated_param.annotation:
             errors.append(
-                f"types for the `{name}` param are do not match; generated={repr(generated_param.annotation)} custom={repr(generated_param.annotation)}"
+                f"types for the `{name}` param are do not match; generated={repr(generated_param.annotation)} custom={repr(custom_param.annotation)}"
             )
             continue