feat: add top_p parameter support to responses API

Author: stainless-app[bot]

.stats.yml

@@ -1,4 +1,4 @@
 configured_endpoints: 108
-openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/llamastack%2Fllama-stack-client-621e8b8ef37d5ebf024fe3bf6a59486a90debf01acca2c9bb4e9032e2dff92d3.yml
-openapi_spec_hash: 51f623cd3ea4addf8f939dd4ef8962c8
+openapi_spec_url: https://storage.googleapis.com/stainless-sdk-openapi-specs/llamastack%2Fllama-stack-client-41c9203afe93dd495edeaad0f9494065f2ae7de5d7148207531eeddf9ed4f11e.yml
+openapi_spec_hash: c47e69115bbf13bdde86a076088fdad9
 config_hash: 6aa61d4143c3e3df785972c0287d1370

src/llama_stack_client/resources/responses/responses.py

@@ -113,6 +113,7 @@ def create(
         text: Optional[response_create_params.Text] | Omit = omit,
         tool_choice: Optional[response_create_params.ToolChoice] | Omit = omit,
         tools: Optional[Iterable[response_create_params.Tool]] | Omit = omit,
+        top_p: Optional[float] | Omit = omit,
         truncation: Optional[Literal["auto", "disabled"]] | Omit = omit,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
@@ -176,6 +177,9 @@ def create(
 
           tools: List of tools available to the model.
 
+          top_p: Nucleus sampling parameter that controls response diversity (lower values
+              increase focus).
+
           truncation: Controls how the service truncates input when it exceeds the model context
               window.
 
@@ -235,6 +239,7 @@ def create(
         text: Optional[response_create_params.Text] | Omit = omit,
         tool_choice: Optional[response_create_params.ToolChoice] | Omit = omit,
         tools: Optional[Iterable[response_create_params.Tool]] | Omit = omit,
+        top_p: Optional[float] | Omit = omit,
         truncation: Optional[Literal["auto", "disabled"]] | Omit = omit,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
@@ -298,6 +303,9 @@ def create(
 
           tools: List of tools available to the model.
 
+          top_p: Nucleus sampling parameter that controls response diversity (lower values
+              increase focus).
+
           truncation: Controls how the service truncates input when it exceeds the model context
               window.
 
@@ -357,6 +365,7 @@ def create(
         text: Optional[response_create_params.Text] | Omit = omit,
         tool_choice: Optional[response_create_params.ToolChoice] | Omit = omit,
         tools: Optional[Iterable[response_create_params.Tool]] | Omit = omit,
+        top_p: Optional[float] | Omit = omit,
         truncation: Optional[Literal["auto", "disabled"]] | Omit = omit,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
@@ -420,6 +429,9 @@ def create(
 
           tools: List of tools available to the model.
 
+          top_p: Nucleus sampling parameter that controls response diversity (lower values
+              increase focus).
+
           truncation: Controls how the service truncates input when it exceeds the model context
               window.
 
@@ -479,6 +491,7 @@ def create(
         text: Optional[response_create_params.Text] | Omit = omit,
         tool_choice: Optional[response_create_params.ToolChoice] | Omit = omit,
         tools: Optional[Iterable[response_create_params.Tool]] | Omit = omit,
+        top_p: Optional[float] | Omit = omit,
         truncation: Optional[Literal["auto", "disabled"]] | Omit = omit,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
@@ -515,6 +528,7 @@ def create(
                     "text": text,
                     "tool_choice": tool_choice,
                     "tools": tools,
+                    "top_p": top_p,
                     "truncation": truncation,
                 },
                 response_create_params.ResponseCreateParamsStreaming
@@ -725,6 +739,7 @@ async def create(
         text: Optional[response_create_params.Text] | Omit = omit,
         tool_choice: Optional[response_create_params.ToolChoice] | Omit = omit,
         tools: Optional[Iterable[response_create_params.Tool]] | Omit = omit,
+        top_p: Optional[float] | Omit = omit,
         truncation: Optional[Literal["auto", "disabled"]] | Omit = omit,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
@@ -788,6 +803,9 @@ async def create(
 
           tools: List of tools available to the model.
 
+          top_p: Nucleus sampling parameter that controls response diversity (lower values
+              increase focus).
+
           truncation: Controls how the service truncates input when it exceeds the model context
               window.
 
@@ -847,6 +865,7 @@ async def create(
         text: Optional[response_create_params.Text] | Omit = omit,
         tool_choice: Optional[response_create_params.ToolChoice] | Omit = omit,
         tools: Optional[Iterable[response_create_params.Tool]] | Omit = omit,
+        top_p: Optional[float] | Omit = omit,
         truncation: Optional[Literal["auto", "disabled"]] | Omit = omit,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
@@ -910,6 +929,9 @@ async def create(
 
           tools: List of tools available to the model.
 
+          top_p: Nucleus sampling parameter that controls response diversity (lower values
+              increase focus).
+
           truncation: Controls how the service truncates input when it exceeds the model context
               window.
 
@@ -969,6 +991,7 @@ async def create(
         text: Optional[response_create_params.Text] | Omit = omit,
         tool_choice: Optional[response_create_params.ToolChoice] | Omit = omit,
         tools: Optional[Iterable[response_create_params.Tool]] | Omit = omit,
+        top_p: Optional[float] | Omit = omit,
         truncation: Optional[Literal["auto", "disabled"]] | Omit = omit,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
@@ -1032,6 +1055,9 @@ async def create(
 
           tools: List of tools available to the model.
 
+          top_p: Nucleus sampling parameter that controls response diversity (lower values
+              increase focus).
+
           truncation: Controls how the service truncates input when it exceeds the model context
               window.
 
@@ -1091,6 +1117,7 @@ async def create(
         text: Optional[response_create_params.Text] | Omit = omit,
         tool_choice: Optional[response_create_params.ToolChoice] | Omit = omit,
         tools: Optional[Iterable[response_create_params.Tool]] | Omit = omit,
+        top_p: Optional[float] | Omit = omit,
         truncation: Optional[Literal["auto", "disabled"]] | Omit = omit,
         # Use the following arguments if you need to pass additional parameters to the API that aren't available via kwargs.
         # The extra values given here take precedence over values defined on the client or passed to this method.
@@ -1127,6 +1154,7 @@ async def create(
                     "text": text,
                     "tool_choice": tool_choice,
                     "tools": tools,
+                    "top_p": top_p,
                     "truncation": truncation,
                 },
                 response_create_params.ResponseCreateParamsStreaming

src/llama_stack_client/types/response_create_params.py

@@ -170,6 +170,12 @@ class ResponseCreateParamsBase(TypedDict, total=False):
     tools: Optional[Iterable[Tool]]
     """List of tools available to the model."""
 
+    top_p: Optional[float]
+    """
+    Nucleus sampling parameter that controls response diversity (lower values
+    increase focus).
+    """
+
     truncation: Optional[Literal["auto", "disabled"]]
     """
     Controls how the service truncates input when it exceeds the model context

tests/api_resources/test_responses.py

@@ -85,6 +85,7 @@ def test_method_create_with_all_params_overload_1(self, client: LlamaStackClient
                     "type": "web_search",
                 }
             ],
+            top_p=0,
             truncation="auto",
         )
         assert_matches_type(ResponseObject, response, path=["response"])
@@ -173,6 +174,7 @@ def test_method_create_with_all_params_overload_2(self, client: LlamaStackClient
                     "type": "web_search",
                 }
             ],
+            top_p=0,
             truncation="auto",
         )
         response_stream.response.close()
@@ -378,6 +380,7 @@ async def test_method_create_with_all_params_overload_1(self, async_client: Asyn
                     "type": "web_search",
                 }
             ],
+            top_p=0,
             truncation="auto",
         )
         assert_matches_type(ResponseObject, response, path=["response"])
@@ -466,6 +469,7 @@ async def test_method_create_with_all_params_overload_2(self, async_client: Asyn
                     "type": "web_search",
                 }
             ],
+            top_p=0,
             truncation="auto",
         )
         await response_stream.response.aclose()