orq-ai-sdk 4.2.0rc28__py3-none-any.whl → 4.2.6__py3-none-any.whl

orq_ai_sdk/models/streamrunagentop.py

@@ -167,6 +167,22 @@ class StreamRunAgentResponseFormatAgentsJSONSchema(BaseModel):
     strict: Optional[bool] = False
     r"""Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is true."""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["description", "schema", "strict"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 class StreamRunAgentResponseFormatJSONSchemaTypedDict(TypedDict):
     r"""
@@ -297,6 +313,22 @@ class StreamRunAgentModelConfigurationStreamOptions(BaseModel):
     include_usage: Optional[bool] = None
     r"""If set, an additional chunk will be streamed before the data: [DONE] message. The usage field on this chunk shows the token usage statistics for the entire request, and the choices field will always be an empty array. All other chunks will also include a usage field, but with a null value."""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["include_usage"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentModelConfigurationThinkingTypedDict = TypeAliasType(
     "StreamRunAgentModelConfigurationThinkingTypedDict",
@@ -339,6 +371,22 @@ class StreamRunAgentToolChoice2(BaseModel):
     type: Optional[StreamRunAgentToolChoiceType] = None
     r"""The type of the tool. Currently, only function is supported."""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["type"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentToolChoice1 = Literal[
     "none",
@@ -405,6 +453,156 @@ class StreamRunAgentModelConfigurationGuardrails(BaseModel):
     r"""Determines whether the guardrail runs on the input (user message) or output (model response)."""
 
 
+class StreamRunAgentModelConfigurationFallbacksTypedDict(TypedDict):
+    model: str
+    r"""Fallback model identifier"""
+
+
+class StreamRunAgentModelConfigurationFallbacks(BaseModel):
+    model: str
+    r"""Fallback model identifier"""
+
+
+class StreamRunAgentModelConfigurationRetryTypedDict(TypedDict):
+    r"""Retry configuration for the request"""
+
+    count: NotRequired[float]
+    r"""Number of retry attempts (1-5)"""
+    on_codes: NotRequired[List[float]]
+    r"""HTTP status codes that trigger retry logic"""
+
+
+class StreamRunAgentModelConfigurationRetry(BaseModel):
+    r"""Retry configuration for the request"""
+
+    count: Optional[float] = 3
+    r"""Number of retry attempts (1-5)"""
+
+    on_codes: Optional[List[float]] = None
+    r"""HTTP status codes that trigger retry logic"""
+
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["count", "on_codes"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
+
+StreamRunAgentModelConfigurationType = Literal["exact_match",]
+
+
+class StreamRunAgentModelConfigurationCacheTypedDict(TypedDict):
+    r"""Cache configuration for the request."""
+
+    type: StreamRunAgentModelConfigurationType
+    ttl: NotRequired[float]
+    r"""Time to live for cached responses in seconds. Maximum 259200 seconds (3 days)."""
+
+
+class StreamRunAgentModelConfigurationCache(BaseModel):
+    r"""Cache configuration for the request."""
+
+    type: StreamRunAgentModelConfigurationType
+
+    ttl: Optional[float] = 1800
+    r"""Time to live for cached responses in seconds. Maximum 259200 seconds (3 days)."""
+
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["ttl"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
+
+StreamRunAgentLoadBalancerType = Literal["weight_based",]
+
+
+class StreamRunAgentLoadBalancerModelsTypedDict(TypedDict):
+    model: str
+    r"""Model identifier for load balancing"""
+    weight: NotRequired[float]
+    r"""Weight assigned to this model for load balancing"""
+
+
+class StreamRunAgentLoadBalancerModels(BaseModel):
+    model: str
+    r"""Model identifier for load balancing"""
+
+    weight: Optional[float] = 0.5
+    r"""Weight assigned to this model for load balancing"""
+
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["weight"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
+
+class StreamRunAgentLoadBalancer1TypedDict(TypedDict):
+    type: StreamRunAgentLoadBalancerType
+    models: List[StreamRunAgentLoadBalancerModelsTypedDict]
+
+
+class StreamRunAgentLoadBalancer1(BaseModel):
+    type: StreamRunAgentLoadBalancerType
+
+    models: List[StreamRunAgentLoadBalancerModels]
+
+
+StreamRunAgentModelConfigurationLoadBalancerTypedDict = (
+    StreamRunAgentLoadBalancer1TypedDict
+)
+r"""Load balancer configuration for the request."""
+
+
+StreamRunAgentModelConfigurationLoadBalancer = StreamRunAgentLoadBalancer1
+r"""Load balancer configuration for the request."""
+
+
+class StreamRunAgentModelConfigurationTimeoutTypedDict(TypedDict):
+    r"""Timeout configuration to apply to the request. If the request exceeds the timeout, it will be retried or fallback to the next model if configured."""
+
+    call_timeout: float
+    r"""Timeout value in milliseconds"""
+
+
+class StreamRunAgentModelConfigurationTimeout(BaseModel):
+    r"""Timeout configuration to apply to the request. If the request exceeds the timeout, it will be retried or fallback to the next model if configured."""
+
+    call_timeout: float
+    r"""Timeout value in milliseconds"""
+
+
 class StreamRunAgentModelConfigurationParametersTypedDict(TypedDict):
     r"""Model behavior parameters that control how the model generates responses. Common parameters: `temperature` (0-1, randomness), `max_completion_tokens` (max output length), `top_p` (sampling diversity). Advanced: `frequency_penalty`, `presence_penalty`, `response_format` (JSON/structured), `reasoning_effort`, `seed` (reproducibility). Support varies by model - consult AI Gateway documentation."""
 
@@ -466,6 +664,16 @@ class StreamRunAgentModelConfigurationParametersTypedDict(TypedDict):
     r"""Output types that you would like the model to generate. Most models are capable of generating text, which is the default: [\"text\"]. The gpt-4o-audio-preview model can also be used to generate audio. To request that this model generate both text and audio responses, you can use: [\"text\", \"audio\"]."""
     guardrails: NotRequired[List[StreamRunAgentModelConfigurationGuardrailsTypedDict]]
     r"""A list of guardrails to apply to the request."""
+    fallbacks: NotRequired[List[StreamRunAgentModelConfigurationFallbacksTypedDict]]
+    r"""Array of fallback models to use if primary model fails"""
+    retry: NotRequired[StreamRunAgentModelConfigurationRetryTypedDict]
+    r"""Retry configuration for the request"""
+    cache: NotRequired[StreamRunAgentModelConfigurationCacheTypedDict]
+    r"""Cache configuration for the request."""
+    load_balancer: NotRequired[StreamRunAgentModelConfigurationLoadBalancerTypedDict]
+    r"""Load balancer configuration for the request."""
+    timeout: NotRequired[StreamRunAgentModelConfigurationTimeoutTypedDict]
+    r"""Timeout configuration to apply to the request. If the request exceeds the timeout, it will be retried or fallback to the next model if configured."""
 
 
 class StreamRunAgentModelConfigurationParameters(BaseModel):
@@ -551,77 +759,96 @@ class StreamRunAgentModelConfigurationParameters(BaseModel):
     guardrails: Optional[List[StreamRunAgentModelConfigurationGuardrails]] = None
     r"""A list of guardrails to apply to the request."""
 
+    fallbacks: Optional[List[StreamRunAgentModelConfigurationFallbacks]] = None
+    r"""Array of fallback models to use if primary model fails"""
+
+    retry: Optional[StreamRunAgentModelConfigurationRetry] = None
+    r"""Retry configuration for the request"""
+
+    cache: Optional[StreamRunAgentModelConfigurationCache] = None
+    r"""Cache configuration for the request."""
+
+    load_balancer: Optional[StreamRunAgentModelConfigurationLoadBalancer] = None
+    r"""Load balancer configuration for the request."""
+
+    timeout: Optional[StreamRunAgentModelConfigurationTimeout] = None
+    r"""Timeout configuration to apply to the request. If the request exceeds the timeout, it will be retried or fallback to the next model if configured."""
+
     @model_serializer(mode="wrap")
     def serialize_model(self, handler):
-        optional_fields = [
-            "audio",
-            "frequency_penalty",
-            "max_tokens",
-            "max_completion_tokens",
-            "logprobs",
-            "top_logprobs",
-            "n",
-            "presence_penalty",
-            "response_format",
-            "reasoning_effort",
-            "verbosity",
-            "seed",
-            "stop",
-            "stream_options",
-            "thinking",
-            "temperature",
-            "top_p",
-            "top_k",
-            "tool_choice",
-            "parallel_tool_calls",
-            "modalities",
-            "guardrails",
-        ]
-        nullable_fields = [
-            "audio",
-            "frequency_penalty",
-            "max_tokens",
-            "max_completion_tokens",
-            "logprobs",
-            "top_logprobs",
-            "n",
-            "presence_penalty",
-            "seed",
-            "stop",
-            "stream_options",
-            "temperature",
-            "top_p",
-            "top_k",
-            "modalities",
-        ]
-        null_default_fields = []
-
+        optional_fields = set(
+            [
+                "audio",
+                "frequency_penalty",
+                "max_tokens",
+                "max_completion_tokens",
+                "logprobs",
+                "top_logprobs",
+                "n",
+                "presence_penalty",
+                "response_format",
+                "reasoning_effort",
+                "verbosity",
+                "seed",
+                "stop",
+                "stream_options",
+                "thinking",
+                "temperature",
+                "top_p",
+                "top_k",
+                "tool_choice",
+                "parallel_tool_calls",
+                "modalities",
+                "guardrails",
+                "fallbacks",
+                "retry",
+                "cache",
+                "load_balancer",
+                "timeout",
+            ]
+        )
+        nullable_fields = set(
+            [
+                "audio",
+                "frequency_penalty",
+                "max_tokens",
+                "max_completion_tokens",
+                "logprobs",
+                "top_logprobs",
+                "n",
+                "presence_penalty",
+                "seed",
+                "stop",
+                "stream_options",
+                "temperature",
+                "top_p",
+                "top_k",
+                "modalities",
+            ]
+        )
         serialized = handler(self)
-
         m = {}
 
         for n, f in type(self).model_fields.items():
             k = f.alias or n
             val = serialized.get(k)
-            serialized.pop(k, None)
-
-            optional_nullable = k in optional_fields and k in nullable_fields
-            is_set = (
-                self.__pydantic_fields_set__.intersection({n})
-                or k in null_default_fields
-            )  # pylint: disable=no-member
-
-            if val is not None and val != UNSET_SENTINEL:
-                m[k] = val
-            elif val != UNSET_SENTINEL and (
-                not k in optional_fields or (optional_nullable and is_set)
-            ):
-                m[k] = val
+            is_nullable_and_explicitly_set = (
+                k in nullable_fields
+                and (self.__pydantic_fields_set__.intersection({n}))  # pylint: disable=no-member
+            )
+
+            if val != UNSET_SENTINEL:
+                if (
+                    val is not None
+                    or k not in optional_fields
+                    or is_nullable_and_explicitly_set
+                ):
+                    m[k] = val
 
         return m
 
 
-class StreamRunAgentModelConfigurationRetryTypedDict(TypedDict):
+class StreamRunAgentModelConfigurationAgentsRetryTypedDict(TypedDict):
     r"""Retry configuration for model requests. Retries are triggered for specific HTTP status codes (e.g., 500, 429, 502, 503, 504). Supports configurable retry count (1-5) and custom status codes."""
 
     count: NotRequired[float]
@@ -630,7 +857,7 @@ class StreamRunAgentModelConfigurationRetryTypedDict(TypedDict):
     r"""HTTP status codes that trigger retry logic"""
 
 
-class StreamRunAgentModelConfigurationRetry(BaseModel):
+class StreamRunAgentModelConfigurationAgentsRetry(BaseModel):
     r"""Retry configuration for model requests. Retries are triggered for specific HTTP status codes (e.g., 500, 429, 502, 503, 504). Supports configurable retry count (1-5) and custom status codes."""
 
     count: Optional[float] = 3
@@ -639,6 +866,22 @@ class StreamRunAgentModelConfigurationRetry(BaseModel):
     on_codes: Optional[List[float]] = None
     r"""HTTP status codes that trigger retry logic"""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["count", "on_codes"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 class StreamRunAgentModelConfiguration2TypedDict(TypedDict):
     r"""
@@ -650,7 +893,7 @@ class StreamRunAgentModelConfiguration2TypedDict(TypedDict):
     r"""A model ID string (e.g., `openai/gpt-4o` or `anthropic/claude-haiku-4-5-20251001`). Only models that support tool calling can be used with agents."""
     parameters: NotRequired[StreamRunAgentModelConfigurationParametersTypedDict]
     r"""Model behavior parameters that control how the model generates responses. Common parameters: `temperature` (0-1, randomness), `max_completion_tokens` (max output length), `top_p` (sampling diversity). Advanced: `frequency_penalty`, `presence_penalty`, `response_format` (JSON/structured), `reasoning_effort`, `seed` (reproducibility). Support varies by model - consult AI Gateway documentation."""
-    retry: NotRequired[StreamRunAgentModelConfigurationRetryTypedDict]
+    retry: NotRequired[StreamRunAgentModelConfigurationAgentsRetryTypedDict]
     r"""Retry configuration for model requests. Retries are triggered for specific HTTP status codes (e.g., 500, 429, 502, 503, 504). Supports configurable retry count (1-5) and custom status codes."""
 
 
@@ -666,9 +909,25 @@ class StreamRunAgentModelConfiguration2(BaseModel):
     parameters: Optional[StreamRunAgentModelConfigurationParameters] = None
     r"""Model behavior parameters that control how the model generates responses. Common parameters: `temperature` (0-1, randomness), `max_completion_tokens` (max output length), `top_p` (sampling diversity). Advanced: `frequency_penalty`, `presence_penalty`, `response_format` (JSON/structured), `reasoning_effort`, `seed` (reproducibility). Support varies by model - consult AI Gateway documentation."""
 
-    retry: Optional[StreamRunAgentModelConfigurationRetry] = None
+    retry: Optional[StreamRunAgentModelConfigurationAgentsRetry] = None
     r"""Retry configuration for model requests. Retries are triggered for specific HTTP status codes (e.g., 500, 429, 502, 503, 504). Supports configurable retry count (1-5) and custom status codes."""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["parameters", "retry"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentModelConfigurationTypedDict = TypeAliasType(
     "StreamRunAgentModelConfigurationTypedDict",
@@ -756,6 +1015,22 @@ class StreamRunAgentResponseFormatAgentsRequestRequestBodyJSONSchema(BaseModel):
     strict: Optional[bool] = False
     r"""Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the schema field. Only a subset of JSON Schema is supported when strict is true."""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["description", "schema", "strict"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 class StreamRunAgentResponseFormatAgentsRequestJSONSchemaTypedDict(TypedDict):
     r"""
@@ -890,6 +1165,22 @@ class StreamRunAgentFallbackModelConfigurationStreamOptions(BaseModel):
     include_usage: Optional[bool] = None
     r"""If set, an additional chunk will be streamed before the data: [DONE] message. The usage field on this chunk shows the token usage statistics for the entire request, and the choices field will always be an empty array. All other chunks will also include a usage field, but with a null value."""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["include_usage"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentFallbackModelConfigurationThinkingTypedDict = TypeAliasType(
     "StreamRunAgentFallbackModelConfigurationThinkingTypedDict",
@@ -932,6 +1223,22 @@ class StreamRunAgentToolChoiceAgents2(BaseModel):
     type: Optional[StreamRunAgentToolChoiceAgentsType] = None
     r"""The type of the tool. Currently, only function is supported."""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["type"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentToolChoiceAgents1 = Literal[
     "none",
@@ -999,66 +1306,216 @@ class StreamRunAgentFallbackModelConfigurationGuardrails(BaseModel):
     r"""Determines whether the guardrail runs on the input (user message) or output (model response)."""
 
 
-class StreamRunAgentFallbackModelConfigurationParametersTypedDict(TypedDict):
-    r"""Optional model parameters specific to this fallback model. Overrides primary model parameters if this fallback is used."""
+class StreamRunAgentFallbackModelConfigurationFallbacksTypedDict(TypedDict):
+    model: str
+    r"""Fallback model identifier"""
 
-    audio: NotRequired[Nullable[StreamRunAgentFallbackModelConfigurationAudioTypedDict]]
-    r"""Parameters for audio output. Required when audio output is requested with modalities: [\"audio\"]. Learn more."""
-    frequency_penalty: NotRequired[Nullable[float]]
-    r"""Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."""
-    max_tokens: NotRequired[Nullable[int]]
-    r"""`[Deprecated]`. The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API.
 
-    This value is now `deprecated` in favor of `max_completion_tokens`, and is not compatible with o1 series models.
-    """
-    max_completion_tokens: NotRequired[Nullable[int]]
-    r"""An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens"""
-    logprobs: NotRequired[Nullable[bool]]
-    r"""Whether to return log probabilities of the output tokens or not. If true, returns the log probabilities of each output token returned in the content of message."""
-    top_logprobs: NotRequired[Nullable[int]]
-    r"""An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to true if this parameter is used."""
-    n: NotRequired[Nullable[int]]
-    r"""How many chat completion choices to generate for each input message. Note that you will be charged based on the number of generated tokens across all of the choices. Keep n as 1 to minimize costs."""
-    presence_penalty: NotRequired[Nullable[float]]
-    r"""Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."""
-    response_format: NotRequired[
-        StreamRunAgentFallbackModelConfigurationResponseFormatTypedDict
-    ]
-    r"""An object specifying the format that the model must output"""
-    reasoning_effort: NotRequired[
-        StreamRunAgentFallbackModelConfigurationReasoningEffort
-    ]
-    r"""Constrains effort on reasoning for [reasoning models](https://platform.openai.com/docs/guides/reasoning). Currently supported values are `none`, `minimal`, `low`, `medium`, `high`, and `xhigh`. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response.
+class StreamRunAgentFallbackModelConfigurationFallbacks(BaseModel):
+    model: str
+    r"""Fallback model identifier"""
 
-    - `gpt-5.1` defaults to `none`, which does not perform reasoning. The supported reasoning values for `gpt-5.1` are `none`, `low`, `medium`, and `high`. Tool calls are supported for all reasoning values in gpt-5.1.
-    - All models before `gpt-5.1` default to `medium` reasoning effort, and do not support `none`.
-    - The `gpt-5-pro` model defaults to (and only supports) `high` reasoning effort.
-    - `xhigh` is currently only supported for `gpt-5.1-codex-max`.
 
-    Any of \"none\", \"minimal\", \"low\", \"medium\", \"high\", \"xhigh\".
-    """
-    verbosity: NotRequired[str]
-    r"""Adjusts response verbosity. Lower levels yield shorter answers."""
-    seed: NotRequired[Nullable[float]]
-    r"""If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."""
-    stop: NotRequired[Nullable[StreamRunAgentFallbackModelConfigurationStopTypedDict]]
-    r"""Up to 4 sequences where the API will stop generating further tokens."""
-    stream_options: NotRequired[
-        Nullable[StreamRunAgentFallbackModelConfigurationStreamOptionsTypedDict]
-    ]
-    r"""Options for streaming response. Only set this when you set stream: true."""
-    thinking: NotRequired[StreamRunAgentFallbackModelConfigurationThinkingTypedDict]
-    temperature: NotRequired[Nullable[float]]
-    r"""What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic."""
-    top_p: NotRequired[Nullable[float]]
-    r"""An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass."""
-    top_k: NotRequired[Nullable[float]]
-    r"""Limits the model to consider only the top k most likely tokens at each step."""
-    tool_choice: NotRequired[
-        StreamRunAgentFallbackModelConfigurationToolChoiceTypedDict
-    ]
-    r"""Controls which (if any) tool is called by the model."""
-    parallel_tool_calls: NotRequired[bool]
+class StreamRunAgentFallbackModelConfigurationRetryTypedDict(TypedDict):
+    r"""Retry configuration for the request"""
+
+    count: NotRequired[float]
+    r"""Number of retry attempts (1-5)"""
+    on_codes: NotRequired[List[float]]
+    r"""HTTP status codes that trigger retry logic"""
+
+
+class StreamRunAgentFallbackModelConfigurationRetry(BaseModel):
+    r"""Retry configuration for the request"""
+
+    count: Optional[float] = 3
+    r"""Number of retry attempts (1-5)"""
+
+    on_codes: Optional[List[float]] = None
+    r"""HTTP status codes that trigger retry logic"""
+
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["count", "on_codes"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
+
+StreamRunAgentFallbackModelConfigurationType = Literal["exact_match",]
+
+
+class StreamRunAgentFallbackModelConfigurationCacheTypedDict(TypedDict):
+    r"""Cache configuration for the request."""
+
+    type: StreamRunAgentFallbackModelConfigurationType
+    ttl: NotRequired[float]
+    r"""Time to live for cached responses in seconds. Maximum 259200 seconds (3 days)."""
+
+
+class StreamRunAgentFallbackModelConfigurationCache(BaseModel):
+    r"""Cache configuration for the request."""
+
+    type: StreamRunAgentFallbackModelConfigurationType
+
+    ttl: Optional[float] = 1800
+    r"""Time to live for cached responses in seconds. Maximum 259200 seconds (3 days)."""
+
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["ttl"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
+
+StreamRunAgentLoadBalancerAgentsType = Literal["weight_based",]
+
+
+class StreamRunAgentLoadBalancerAgentsModelsTypedDict(TypedDict):
+    model: str
+    r"""Model identifier for load balancing"""
+    weight: NotRequired[float]
+    r"""Weight assigned to this model for load balancing"""
+
+
+class StreamRunAgentLoadBalancerAgentsModels(BaseModel):
+    model: str
+    r"""Model identifier for load balancing"""
+
+    weight: Optional[float] = 0.5
+    r"""Weight assigned to this model for load balancing"""
+
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["weight"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
+
+class StreamRunAgentLoadBalancerAgents1TypedDict(TypedDict):
+    type: StreamRunAgentLoadBalancerAgentsType
+    models: List[StreamRunAgentLoadBalancerAgentsModelsTypedDict]
+
+
+class StreamRunAgentLoadBalancerAgents1(BaseModel):
+    type: StreamRunAgentLoadBalancerAgentsType
+
+    models: List[StreamRunAgentLoadBalancerAgentsModels]
+
+
+StreamRunAgentFallbackModelConfigurationLoadBalancerTypedDict = (
+    StreamRunAgentLoadBalancerAgents1TypedDict
+)
+r"""Load balancer configuration for the request."""
+
+
+StreamRunAgentFallbackModelConfigurationLoadBalancer = StreamRunAgentLoadBalancerAgents1
+r"""Load balancer configuration for the request."""
+
+
+class StreamRunAgentFallbackModelConfigurationTimeoutTypedDict(TypedDict):
+    r"""Timeout configuration to apply to the request. If the request exceeds the timeout, it will be retried or fallback to the next model if configured."""
+
+    call_timeout: float
+    r"""Timeout value in milliseconds"""
+
+
+class StreamRunAgentFallbackModelConfigurationTimeout(BaseModel):
+    r"""Timeout configuration to apply to the request. If the request exceeds the timeout, it will be retried or fallback to the next model if configured."""
+
+    call_timeout: float
+    r"""Timeout value in milliseconds"""
+
+
+class StreamRunAgentFallbackModelConfigurationParametersTypedDict(TypedDict):
+    r"""Optional model parameters specific to this fallback model. Overrides primary model parameters if this fallback is used."""
+
+    audio: NotRequired[Nullable[StreamRunAgentFallbackModelConfigurationAudioTypedDict]]
+    r"""Parameters for audio output. Required when audio output is requested with modalities: [\"audio\"]. Learn more."""
+    frequency_penalty: NotRequired[Nullable[float]]
+    r"""Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim."""
+    max_tokens: NotRequired[Nullable[int]]
+    r"""`[Deprecated]`. The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API.
+
+    This value is now `deprecated` in favor of `max_completion_tokens`, and is not compatible with o1 series models.
+    """
+    max_completion_tokens: NotRequired[Nullable[int]]
+    r"""An upper bound for the number of tokens that can be generated for a completion, including visible output tokens and reasoning tokens"""
+    logprobs: NotRequired[Nullable[bool]]
+    r"""Whether to return log probabilities of the output tokens or not. If true, returns the log probabilities of each output token returned in the content of message."""
+    top_logprobs: NotRequired[Nullable[int]]
+    r"""An integer between 0 and 20 specifying the number of most likely tokens to return at each token position, each with an associated log probability. logprobs must be set to true if this parameter is used."""
+    n: NotRequired[Nullable[int]]
+    r"""How many chat completion choices to generate for each input message. Note that you will be charged based on the number of generated tokens across all of the choices. Keep n as 1 to minimize costs."""
+    presence_penalty: NotRequired[Nullable[float]]
+    r"""Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics."""
+    response_format: NotRequired[
+        StreamRunAgentFallbackModelConfigurationResponseFormatTypedDict
+    ]
+    r"""An object specifying the format that the model must output"""
+    reasoning_effort: NotRequired[
+        StreamRunAgentFallbackModelConfigurationReasoningEffort
+    ]
+    r"""Constrains effort on reasoning for [reasoning models](https://platform.openai.com/docs/guides/reasoning). Currently supported values are `none`, `minimal`, `low`, `medium`, `high`, and `xhigh`. Reducing reasoning effort can result in faster responses and fewer tokens used on reasoning in a response.
+
+    - `gpt-5.1` defaults to `none`, which does not perform reasoning. The supported reasoning values for `gpt-5.1` are `none`, `low`, `medium`, and `high`. Tool calls are supported for all reasoning values in gpt-5.1.
+    - All models before `gpt-5.1` default to `medium` reasoning effort, and do not support `none`.
+    - The `gpt-5-pro` model defaults to (and only supports) `high` reasoning effort.
+    - `xhigh` is currently only supported for `gpt-5.1-codex-max`.
+
+    Any of \"none\", \"minimal\", \"low\", \"medium\", \"high\", \"xhigh\".
+    """
+    verbosity: NotRequired[str]
+    r"""Adjusts response verbosity. Lower levels yield shorter answers."""
+    seed: NotRequired[Nullable[float]]
+    r"""If specified, our system will make a best effort to sample deterministically, such that repeated requests with the same seed and parameters should return the same result."""
+    stop: NotRequired[Nullable[StreamRunAgentFallbackModelConfigurationStopTypedDict]]
+    r"""Up to 4 sequences where the API will stop generating further tokens."""
+    stream_options: NotRequired[
+        Nullable[StreamRunAgentFallbackModelConfigurationStreamOptionsTypedDict]
+    ]
+    r"""Options for streaming response. Only set this when you set stream: true."""
+    thinking: NotRequired[StreamRunAgentFallbackModelConfigurationThinkingTypedDict]
+    temperature: NotRequired[Nullable[float]]
+    r"""What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic."""
+    top_p: NotRequired[Nullable[float]]
+    r"""An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass."""
+    top_k: NotRequired[Nullable[float]]
+    r"""Limits the model to consider only the top k most likely tokens at each step."""
+    tool_choice: NotRequired[
+        StreamRunAgentFallbackModelConfigurationToolChoiceTypedDict
+    ]
+    r"""Controls which (if any) tool is called by the model."""
+    parallel_tool_calls: NotRequired[bool]
     r"""Whether to enable parallel function calling during tool use."""
     modalities: NotRequired[
         Nullable[List[StreamRunAgentFallbackModelConfigurationModalities]]
@@ -1068,6 +1525,20 @@ class StreamRunAgentFallbackModelConfigurationParametersTypedDict(TypedDict):
         List[StreamRunAgentFallbackModelConfigurationGuardrailsTypedDict]
     ]
     r"""A list of guardrails to apply to the request."""
+    fallbacks: NotRequired[
+        List[StreamRunAgentFallbackModelConfigurationFallbacksTypedDict]
+    ]
+    r"""Array of fallback models to use if primary model fails"""
+    retry: NotRequired[StreamRunAgentFallbackModelConfigurationRetryTypedDict]
+    r"""Retry configuration for the request"""
+    cache: NotRequired[StreamRunAgentFallbackModelConfigurationCacheTypedDict]
+    r"""Cache configuration for the request."""
+    load_balancer: NotRequired[
+        StreamRunAgentFallbackModelConfigurationLoadBalancerTypedDict
+    ]
+    r"""Load balancer configuration for the request."""
+    timeout: NotRequired[StreamRunAgentFallbackModelConfigurationTimeoutTypedDict]
+    r"""Timeout configuration to apply to the request. If the request exceeds the timeout, it will be retried or fallback to the next model if configured."""
 
 
 class StreamRunAgentFallbackModelConfigurationParameters(BaseModel):
@@ -1159,77 +1630,96 @@ class StreamRunAgentFallbackModelConfigurationParameters(BaseModel):
     )
     r"""A list of guardrails to apply to the request."""
 
+    fallbacks: Optional[List[StreamRunAgentFallbackModelConfigurationFallbacks]] = None
+    r"""Array of fallback models to use if primary model fails"""
+
+    retry: Optional[StreamRunAgentFallbackModelConfigurationRetry] = None
+    r"""Retry configuration for the request"""
+
+    cache: Optional[StreamRunAgentFallbackModelConfigurationCache] = None
+    r"""Cache configuration for the request."""
+
+    load_balancer: Optional[StreamRunAgentFallbackModelConfigurationLoadBalancer] = None
+    r"""Load balancer configuration for the request."""
+
+    timeout: Optional[StreamRunAgentFallbackModelConfigurationTimeout] = None
+    r"""Timeout configuration to apply to the request. If the request exceeds the timeout, it will be retried or fallback to the next model if configured."""
+
     @model_serializer(mode="wrap")
     def serialize_model(self, handler):
-        optional_fields = [
-            "audio",
-            "frequency_penalty",
-            "max_tokens",
-            "max_completion_tokens",
-            "logprobs",
-            "top_logprobs",
-            "n",
-            "presence_penalty",
-            "response_format",
-            "reasoning_effort",
-            "verbosity",
-            "seed",
-            "stop",
-            "stream_options",
-            "thinking",
-            "temperature",
-            "top_p",
-            "top_k",
-            "tool_choice",
-            "parallel_tool_calls",
-            "modalities",
-            "guardrails",
-        ]
-        nullable_fields = [
-            "audio",
-            "frequency_penalty",
-            "max_tokens",
-            "max_completion_tokens",
-            "logprobs",
-            "top_logprobs",
-            "n",
-            "presence_penalty",
-            "seed",
-            "stop",
-            "stream_options",
-            "temperature",
-            "top_p",
-            "top_k",
-            "modalities",
-        ]
-        null_default_fields = []
-
+        optional_fields = set(
+            [
+                "audio",
+                "frequency_penalty",
+                "max_tokens",
+                "max_completion_tokens",
+                "logprobs",
+                "top_logprobs",
+                "n",
+                "presence_penalty",
+                "response_format",
+                "reasoning_effort",
+                "verbosity",
+                "seed",
+                "stop",
+                "stream_options",
+                "thinking",
+                "temperature",
+                "top_p",
+                "top_k",
+                "tool_choice",
+                "parallel_tool_calls",
+                "modalities",
+                "guardrails",
+                "fallbacks",
+                "retry",
+                "cache",
+                "load_balancer",
+                "timeout",
+            ]
+        )
+        nullable_fields = set(
+            [
+                "audio",
+                "frequency_penalty",
+                "max_tokens",
+                "max_completion_tokens",
+                "logprobs",
+                "top_logprobs",
+                "n",
+                "presence_penalty",
+                "seed",
+                "stop",
+                "stream_options",
+                "temperature",
+                "top_p",
+                "top_k",
+                "modalities",
+            ]
+        )
         serialized = handler(self)
-
         m = {}
 
         for n, f in type(self).model_fields.items():
             k = f.alias or n
             val = serialized.get(k)
-            serialized.pop(k, None)
-
-            optional_nullable = k in optional_fields and k in nullable_fields
-            is_set = (
-                self.__pydantic_fields_set__.intersection({n})
-                or k in null_default_fields
-            )  # pylint: disable=no-member
-
-            if val is not None and val != UNSET_SENTINEL:
-                m[k] = val
-            elif val != UNSET_SENTINEL and (
-                not k in optional_fields or (optional_nullable and is_set)
-            ):
-                m[k] = val
+            is_nullable_and_explicitly_set = (
+                k in nullable_fields
+                and (self.__pydantic_fields_set__.intersection({n}))  # pylint: disable=no-member
+            )
+
+            if val != UNSET_SENTINEL:
+                if (
+                    val is not None
+                    or k not in optional_fields
+                    or is_nullable_and_explicitly_set
+                ):
+                    m[k] = val
 
         return m
 
 
-class StreamRunAgentFallbackModelConfigurationRetryTypedDict(TypedDict):
+class StreamRunAgentFallbackModelConfigurationAgentsRetryTypedDict(TypedDict):
     r"""Retry configuration for this fallback model. Allows customizing retry count (1-5) and HTTP status codes that trigger retries."""
 
     count: NotRequired[float]
@@ -1238,7 +1728,7 @@ class StreamRunAgentFallbackModelConfigurationRetryTypedDict(TypedDict):
     r"""HTTP status codes that trigger retry logic"""
 
 
-class StreamRunAgentFallbackModelConfigurationRetry(BaseModel):
+class StreamRunAgentFallbackModelConfigurationAgentsRetry(BaseModel):
     r"""Retry configuration for this fallback model. Allows customizing retry count (1-5) and HTTP status codes that trigger retries."""
 
     count: Optional[float] = 3
@@ -1247,6 +1737,22 @@ class StreamRunAgentFallbackModelConfigurationRetry(BaseModel):
     on_codes: Optional[List[float]] = None
     r"""HTTP status codes that trigger retry logic"""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["count", "on_codes"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 class StreamRunAgentFallbackModelConfiguration2TypedDict(TypedDict):
     r"""Fallback model configuration with optional parameters and retry settings."""
@@ -1255,7 +1761,7 @@ class StreamRunAgentFallbackModelConfiguration2TypedDict(TypedDict):
     r"""A fallback model ID string. Must support tool calling."""
     parameters: NotRequired[StreamRunAgentFallbackModelConfigurationParametersTypedDict]
     r"""Optional model parameters specific to this fallback model. Overrides primary model parameters if this fallback is used."""
-    retry: NotRequired[StreamRunAgentFallbackModelConfigurationRetryTypedDict]
+    retry: NotRequired[StreamRunAgentFallbackModelConfigurationAgentsRetryTypedDict]
     r"""Retry configuration for this fallback model. Allows customizing retry count (1-5) and HTTP status codes that trigger retries."""
 
 
@@ -1268,9 +1774,25 @@ class StreamRunAgentFallbackModelConfiguration2(BaseModel):
     parameters: Optional[StreamRunAgentFallbackModelConfigurationParameters] = None
     r"""Optional model parameters specific to this fallback model. Overrides primary model parameters if this fallback is used."""
 
-    retry: Optional[StreamRunAgentFallbackModelConfigurationRetry] = None
+    retry: Optional[StreamRunAgentFallbackModelConfigurationAgentsRetry] = None
     r"""Retry configuration for this fallback model. Allows customizing retry count (1-5) and HTTP status codes that trigger retries."""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["parameters", "retry"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentFallbackModelConfigurationTypedDict = TypeAliasType(
     "StreamRunAgentFallbackModelConfigurationTypedDict",
@@ -1355,6 +1877,22 @@ class StreamRunAgentA2AMessage(BaseModel):
     message_id: Annotated[Optional[str], pydantic.Field(alias="messageId")] = None
     r"""Optional A2A message ID in ULID format"""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["messageId"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 class StreamRunAgentIdentityTypedDict(TypedDict):
     r"""Information about the identity making the request. If the identity does not exist, it will be created automatically."""
@@ -1394,6 +1932,22 @@ class StreamRunAgentIdentity(BaseModel):
     tags: Optional[List[str]] = None
     r"""A list of tags associated with the contact"""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["display_name", "email", "metadata", "logo_url", "tags"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 @deprecated(
     "warning: ** DEPRECATED ** - This will be removed in a future release, please migrate away from it as soon as possible."
@@ -1439,6 +1993,22 @@ class StreamRunAgentContact(BaseModel):
     tags: Optional[List[str]] = None
     r"""A list of tags associated with the contact"""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["display_name", "email", "metadata", "logo_url", "tags"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 class StreamRunAgentThreadTypedDict(TypedDict):
     r"""Thread information to group related requests"""
@@ -1458,6 +2028,22 @@ class StreamRunAgentThread(BaseModel):
     tags: Optional[List[str]] = None
     r"""Optional tags to differentiate or categorize threads"""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["tags"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 class StreamRunAgentMemoryTypedDict(TypedDict):
     r"""Memory configuration for the agent execution. Used to associate memory stores with specific entities like users or sessions."""
@@ -1497,8 +2083,24 @@ class StreamRunAgentTeamOfAgents(BaseModel):
     role: Optional[str] = None
     r"""The role of the agent in this context. This is used to give extra information to the leader to help it decide which agent to hand off to."""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["role"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
 
-StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools15Type = Literal[
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
+
+StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools16Type = Literal[
     "mcp",
 ]
 
@@ -1513,29 +2115,61 @@ class StreamRunAgentAgentToolInputRunAgentsHeaders(BaseModel):
 
     encrypted: Optional[bool] = False
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["encrypted"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
-StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools15McpType = Literal[
+StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools16McpType = Literal[
     "object",
 ]
 
 
-class AgentToolInputRunSchemaTypedDict(TypedDict):
-    type: StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools15McpType
+class StreamRunAgentAgentToolInputRunAgentsSchemaTypedDict(TypedDict):
+    type: StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools16McpType
     properties: NotRequired[Dict[str, Any]]
     required: NotRequired[List[str]]
 
 
-class AgentToolInputRunSchema(BaseModel):
-    type: StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools15McpType
+class StreamRunAgentAgentToolInputRunAgentsSchema(BaseModel):
+    type: StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools16McpType
 
     properties: Optional[Dict[str, Any]] = None
 
     required: Optional[List[str]] = None
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["properties", "required"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 class AgentToolInputRunToolsTypedDict(TypedDict):
     name: str
-    schema_: AgentToolInputRunSchemaTypedDict
+    schema_: StreamRunAgentAgentToolInputRunAgentsSchemaTypedDict
     id: NotRequired[str]
     description: NotRequired[str]
 
@@ -1543,12 +2177,30 @@ class AgentToolInputRunToolsTypedDict(TypedDict):
 class AgentToolInputRunTools(BaseModel):
     name: str
 
-    schema_: Annotated[AgentToolInputRunSchema, pydantic.Field(alias="schema")]
+    schema_: Annotated[
+        StreamRunAgentAgentToolInputRunAgentsSchema, pydantic.Field(alias="schema")
+    ]
 
-    id: Optional[str] = "01KEXRJ7NDSYG7V3YEZKK55TE9"
+    id: Optional[str] = "01KFTTTR0XG9Q2WCVPG1AD9NB7"
 
     description: Optional[str] = None
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["id", "description"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 AgentToolInputRunConnectionType = Literal[
     "http",
@@ -1583,11 +2235,27 @@ class AgentToolInputRunMcp(BaseModel):
     headers: Optional[Dict[str, StreamRunAgentAgentToolInputRunAgentsHeaders]] = None
     r"""HTTP headers for MCP server requests with encryption support"""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["headers"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 class AgentToolInputRunMCPToolRunTypedDict(TypedDict):
     r"""MCP tool with inline definition for on-the-fly creation in run endpoint"""
 
-    type: StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools15Type
+    type: StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools16Type
     key: str
     r"""Unique key of the tool as it will be displayed in the UI"""
     description: str
@@ -1601,7 +2269,7 @@ class AgentToolInputRunMCPToolRunTypedDict(TypedDict):
 class AgentToolInputRunMCPToolRun(BaseModel):
     r"""MCP tool with inline definition for on-the-fly creation in run endpoint"""
 
-    type: StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools15Type
+    type: StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools16Type
 
     key: str
     r"""Unique key of the tool as it will be displayed in the UI"""
@@ -1617,6 +2285,157 @@ class AgentToolInputRunMCPToolRun(BaseModel):
 
     requires_approval: Optional[bool] = False
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["_id", "display_name", "requires_approval"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
+
+StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools15Type = Literal[
+    "json_schema",
+]
+
+
+class StreamRunAgentAgentToolInputRunSchemaTypedDict(TypedDict):
+    r"""The schema for the response format, described as a JSON Schema object. See the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format."""
+
+    type: str
+    r"""The JSON Schema type"""
+    properties: Dict[str, Any]
+    r"""The properties of the JSON Schema object"""
+    required: List[str]
+    r"""Array of required property names"""
+
+
+class StreamRunAgentAgentToolInputRunSchema(BaseModel):
+    r"""The schema for the response format, described as a JSON Schema object. See the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format."""
+
+    model_config = ConfigDict(
+        populate_by_name=True, arbitrary_types_allowed=True, extra="allow"
+    )
+    __pydantic_extra__: Dict[str, Any] = pydantic.Field(init=False)
+
+    type: str
+    r"""The JSON Schema type"""
+
+    properties: Dict[str, Any]
+    r"""The properties of the JSON Schema object"""
+
+    required: List[str]
+    r"""Array of required property names"""
+
+    @property
+    def additional_properties(self):
+        return self.__pydantic_extra__
+
+    @additional_properties.setter
+    def additional_properties(self, value):
+        self.__pydantic_extra__ = value  # pyright: ignore[reportIncompatibleVariableOverride]
+
+
+class StreamRunAgentAgentToolInputRunJSONSchemaTypedDict(TypedDict):
+    name: str
+    r"""The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."""
+    description: str
+    r"""A description of what the response format is for. This will be shown to the user."""
+    schema_: StreamRunAgentAgentToolInputRunSchemaTypedDict
+    r"""The schema for the response format, described as a JSON Schema object. See the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format."""
+    strict: NotRequired[bool]
+    r"""Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the `schema` field. Only a subset of JSON Schema is supported when `strict` is `true`. Only compatible with `OpenAI` models."""
+
+
+class StreamRunAgentAgentToolInputRunJSONSchema(BaseModel):
+    name: str
+    r"""The name of the response format. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64."""
+
+    description: str
+    r"""A description of what the response format is for. This will be shown to the user."""
+
+    schema_: Annotated[
+        StreamRunAgentAgentToolInputRunSchema, pydantic.Field(alias="schema")
+    ]
+    r"""The schema for the response format, described as a JSON Schema object. See the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format."""
+
+    strict: Optional[bool] = None
+    r"""Whether to enable strict schema adherence when generating the output. If set to true, the model will always follow the exact schema defined in the `schema` field. Only a subset of JSON Schema is supported when `strict` is `true`. Only compatible with `OpenAI` models."""
+
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["strict"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
+
+class AgentToolInputRunJSONSchemaToolRunTypedDict(TypedDict):
+    r"""JSON Schema tool with inline definition for on-the-fly creation in run endpoint"""
+
+    type: StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools15Type
+    key: str
+    r"""Unique key of the tool as it will be displayed in the UI"""
+    description: str
+    r"""A description of the tool, used by the model to choose when and how to call the tool. We do recommend using the `description` field as accurate as possible to give enough context to the model to make the right decision."""
+    json_schema: StreamRunAgentAgentToolInputRunJSONSchemaTypedDict
+    id: NotRequired[str]
+    display_name: NotRequired[str]
+    requires_approval: NotRequired[bool]
+
+
+class AgentToolInputRunJSONSchemaToolRun(BaseModel):
+    r"""JSON Schema tool with inline definition for on-the-fly creation in run endpoint"""
+
+    type: StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools15Type
+
+    key: str
+    r"""Unique key of the tool as it will be displayed in the UI"""
+
+    description: str
+    r"""A description of the tool, used by the model to choose when and how to call the tool. We do recommend using the `description` field as accurate as possible to give enough context to the model to make the right decision."""
+
+    json_schema: StreamRunAgentAgentToolInputRunJSONSchema
+
+    id: Annotated[Optional[str], pydantic.Field(alias="_id")] = None
+
+    display_name: Optional[str] = None
+
+    requires_approval: Optional[bool] = False
+
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["_id", "display_name", "requires_approval"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools14Type = Literal[
     "function",
@@ -1690,6 +2509,22 @@ class StreamRunAgentAgentToolInputRunFunction(BaseModel):
     parameters: Optional[StreamRunAgentAgentToolInputRunAgentsParameters] = None
     r"""The parameters the functions accepts, described as a JSON Schema object. See the `OpenAI` [guide](https://platform.openai.com/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format."""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["description", "strict", "parameters"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 class AgentToolInputRunFunctionToolRunTypedDict(TypedDict):
     r"""Function tool with inline definition for on-the-fly creation in run endpoint"""
@@ -1722,6 +2557,24 @@ class AgentToolInputRunFunctionToolRun(BaseModel):
 
     requires_approval: Optional[bool] = False
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(
+            ["_id", "display_name", "description", "requires_approval"]
+        )
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools13Type = Literal[
     "code",
@@ -1788,8 +2641,24 @@ class AgentToolInputRunCodeTool(BaseModel):
     code: str
     r"""The code to execute."""
 
-    parameters: Optional[StreamRunAgentAgentToolInputRunParameters] = None
-    r"""The parameters the functions accepts, described as a JSON Schema object. See the `OpenAI` [guide](https://platform.openai.com/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format."""
+    parameters: Optional[StreamRunAgentAgentToolInputRunParameters] = None
+    r"""The parameters the functions accepts, described as a JSON Schema object. See the `OpenAI` [guide](https://platform.openai.com/docs/guides/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format."""
+
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["parameters"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
 
 
 class AgentToolInputRunCodeToolRunTypedDict(TypedDict):
@@ -1825,6 +2694,22 @@ class AgentToolInputRunCodeToolRun(BaseModel):
 
     requires_approval: Optional[bool] = False
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["_id", "display_name", "requires_approval"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools12Type = Literal[
     "http",
@@ -1850,6 +2735,22 @@ class StreamRunAgentHeaders2(BaseModel):
 
     encrypted: Optional[bool] = False
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["encrypted"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentAgentToolInputRunHeadersTypedDict = TypeAliasType(
     "StreamRunAgentAgentToolInputRunHeadersTypedDict",
@@ -1890,6 +2791,22 @@ class AgentToolInputRunBlueprint(BaseModel):
     body: Optional[Dict[str, Any]] = None
     r"""The body to send with the request."""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["headers", "body"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools12HTTPType = (
     Literal[
@@ -1937,6 +2854,22 @@ class AgentToolInputRunArguments(BaseModel):
     default_value: Optional[AgentToolInputRunDefaultValue] = None
     r"""The default value of the argument."""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["send_to_model", "default_value"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 class AgentToolInputRunHTTPTypedDict(TypedDict):
     blueprint: AgentToolInputRunBlueprintTypedDict
@@ -1952,6 +2885,22 @@ class AgentToolInputRunHTTP(BaseModel):
     arguments: Optional[Dict[str, AgentToolInputRunArguments]] = None
     r"""The arguments to send with the request. The keys will be used to replace the placeholders in the `blueprint` field."""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["arguments"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 class AgentToolInputRunHTTPToolRunTypedDict(TypedDict):
     r"""HTTP tool with inline definition for on-the-fly creation in run endpoint"""
@@ -1986,6 +2935,22 @@ class AgentToolInputRunHTTPToolRun(BaseModel):
 
     requires_approval: Optional[bool] = False
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["_id", "display_name", "requires_approval"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools11Type = Literal[
     "current_date",
@@ -2008,6 +2973,22 @@ class StreamRunAgentAgentToolInputRunCurrentDateTool(BaseModel):
     requires_approval: Optional[bool] = None
     r"""Whether this tool requires approval before execution"""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["requires_approval"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools10Type = Literal[
     "query_knowledge_base",
@@ -2030,6 +3011,22 @@ class StreamRunAgentAgentToolInputRunQueryKnowledgeBaseTool(BaseModel):
     requires_approval: Optional[bool] = None
     r"""Whether this tool requires approval before execution"""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["requires_approval"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools9Type = Literal[
     "retrieve_knowledge_bases",
@@ -2052,6 +3049,22 @@ class StreamRunAgentAgentToolInputRunRetrieveKnowledgeBasesTool(BaseModel):
     requires_approval: Optional[bool] = None
     r"""Whether this tool requires approval before execution"""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["requires_approval"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools8Type = Literal[
     "delete_memory_document",
@@ -2074,6 +3087,22 @@ class StreamRunAgentAgentToolInputRunDeleteMemoryDocumentTool(BaseModel):
     requires_approval: Optional[bool] = None
     r"""Whether this tool requires approval before execution"""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["requires_approval"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsTools7Type = Literal[
     "retrieve_memory_stores",
@@ -2096,6 +3125,22 @@ class StreamRunAgentAgentToolInputRunRetrieveMemoryStoresTool(BaseModel):
     requires_approval: Optional[bool] = None
     r"""Whether this tool requires approval before execution"""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["requires_approval"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsToolsType = Literal[
     "write_memory_store",
@@ -2118,6 +3163,22 @@ class StreamRunAgentAgentToolInputRunWriteMemoryStoreTool(BaseModel):
     requires_approval: Optional[bool] = None
     r"""Whether this tool requires approval before execution"""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["requires_approval"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentAgentToolInputRunAgentsRequestRequestBodySettingsType = Literal[
     "query_memory_store",
@@ -2140,6 +3201,22 @@ class StreamRunAgentAgentToolInputRunQueryMemoryStoreTool(BaseModel):
     requires_approval: Optional[bool] = None
     r"""Whether this tool requires approval before execution"""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["requires_approval"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentAgentToolInputRunAgentsRequestRequestBodyType = Literal[
     "retrieve_agents",
@@ -2162,6 +3239,22 @@ class StreamRunAgentAgentToolInputRunRetrieveAgentsTool(BaseModel):
     requires_approval: Optional[bool] = None
     r"""Whether this tool requires approval before execution"""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["requires_approval"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentAgentToolInputRunAgentsRequestType = Literal["call_sub_agent",]
 
@@ -2182,6 +3275,22 @@ class StreamRunAgentAgentToolInputRunCallSubAgentTool(BaseModel):
     requires_approval: Optional[bool] = None
     r"""Whether this tool requires approval before execution"""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["requires_approval"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentAgentToolInputRunAgentsType = Literal["web_scraper",]
 
@@ -2202,6 +3311,22 @@ class StreamRunAgentAgentToolInputRunWebScraperTool(BaseModel):
     requires_approval: Optional[bool] = None
     r"""Whether this tool requires approval before execution"""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["requires_approval"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentAgentToolInputRunType = Literal["google_search",]
 
@@ -2222,6 +3347,22 @@ class StreamRunAgentAgentToolInputRunGoogleSearchTool(BaseModel):
     requires_approval: Optional[bool] = None
     r"""Whether this tool requires approval before execution"""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["requires_approval"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentAgentToolInputRunTypedDict = TypeAliasType(
     "StreamRunAgentAgentToolInputRunTypedDict",
@@ -2240,10 +3381,11 @@ StreamRunAgentAgentToolInputRunTypedDict = TypeAliasType(
         AgentToolInputRunHTTPToolRunTypedDict,
         AgentToolInputRunCodeToolRunTypedDict,
         AgentToolInputRunFunctionToolRunTypedDict,
+        AgentToolInputRunJSONSchemaToolRunTypedDict,
         AgentToolInputRunMCPToolRunTypedDict,
     ],
 )
-r"""Tool configuration for agent run operations. Built-in tools only require a type and requires_approval, while custom tools (HTTP, Code, Function, MCP) support full inline definitions for on-the-fly creation."""
+r"""Tool configuration for agent run operations. Built-in tools only require a type and requires_approval, while custom tools (HTTP, Code, Function, JSON Schema, MCP) support full inline definitions for on-the-fly creation."""
 
 
 StreamRunAgentAgentToolInputRun = Annotated[
@@ -2286,11 +3428,12 @@ StreamRunAgentAgentToolInputRun = Annotated[
         Annotated[AgentToolInputRunHTTPToolRun, Tag("http")],
         Annotated[AgentToolInputRunCodeToolRun, Tag("code")],
         Annotated[AgentToolInputRunFunctionToolRun, Tag("function")],
+        Annotated[AgentToolInputRunJSONSchemaToolRun, Tag("json_schema")],
         Annotated[AgentToolInputRunMCPToolRun, Tag("mcp")],
     ],
     Discriminator(lambda m: get_discriminator(m, "type", "type")),
 ]
-r"""Tool configuration for agent run operations. Built-in tools only require a type and requires_approval, while custom tools (HTTP, Code, Function, MCP) support full inline definitions for on-the-fly creation."""
+r"""Tool configuration for agent run operations. Built-in tools only require a type and requires_approval, while custom tools (HTTP, Code, Function, JSON Schema, MCP) support full inline definitions for on-the-fly creation."""
 
 
 StreamRunAgentToolApprovalRequired = Literal[
@@ -2327,6 +3470,22 @@ class StreamRunAgentEvaluators(BaseModel):
     sample_rate: Optional[float] = 50
     r"""The percentage of executions to evaluate with this evaluator (1-100). For example, a value of 50 means the evaluator will run on approximately half of the executions."""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["sample_rate"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 StreamRunAgentAgentsExecuteOn = Literal[
     "input",
@@ -2354,6 +3513,22 @@ class StreamRunAgentGuardrails(BaseModel):
     sample_rate: Optional[float] = 50
     r"""The percentage of executions to evaluate with this evaluator (1-100). For example, a value of 50 means the evaluator will run on approximately half of the executions."""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(["sample_rate"])
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 class StreamRunAgentSettingsTypedDict(TypedDict):
     tools: NotRequired[List[StreamRunAgentAgentToolInputRunTypedDict]]
@@ -2389,6 +3564,31 @@ class StreamRunAgentSettings(BaseModel):
     guardrails: Optional[List[StreamRunAgentGuardrails]] = None
     r"""Configuration for a guardrail applied to the agent"""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(
+            [
+                "tools",
+                "tool_approval_required",
+                "max_iterations",
+                "max_execution_time",
+                "evaluators",
+                "guardrails",
+            ]
+        )
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 class StreamRunAgentRequestBodyTypedDict(TypedDict):
     key: str
@@ -2514,6 +3714,39 @@ class StreamRunAgentRequestBody(BaseModel):
     stream_timeout_seconds: Optional[float] = None
     r"""Stream timeout in seconds (1-3600). Default: 1800 (30 minutes)"""
 
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = set(
+            [
+                "task_id",
+                "fallback_models",
+                "variables",
+                "identity",
+                "contact",
+                "thread",
+                "memory",
+                "description",
+                "system_prompt",
+                "memory_stores",
+                "knowledge_bases",
+                "team_of_agents",
+                "metadata",
+                "stream_timeout_seconds",
+            ]
+        )
+        serialized = handler(self)
+        m = {}
+
+        for n, f in type(self).model_fields.items():
+            k = f.alias or n
+            val = serialized.get(k)
+
+            if val != UNSET_SENTINEL:
+                if val is not None or k not in optional_fields:
+                    m[k] = val
+
+        return m
+
 
 class StreamRunAgentAgentsResponseBodyData(BaseModel):
     message: str