openrouter 0.0.22__py3-none-any.whl → 0.1.2__py3-none-any.whl

openrouter/components/openresponsesrequest.py

@@ -33,9 +33,18 @@ from .openresponseswebsearchtool import (
     OpenResponsesWebSearchTool,
     OpenResponsesWebSearchToolTypedDict,
 )
+from .pdfparseroptions import PDFParserOptions, PDFParserOptionsTypedDict
+from .preferredmaxlatency import PreferredMaxLatency, PreferredMaxLatencyTypedDict
+from .preferredminthroughput import (
+    PreferredMinThroughput,
+    PreferredMinThroughputTypedDict,
+)
 from .providername import ProviderName
 from .providersort import ProviderSort
+from .providersortconfig import ProviderSortConfig, ProviderSortConfigTypedDict
 from .quantization import Quantization
+from .responsesoutputmodality import ResponsesOutputModality
+from .websearchengine import WebSearchEngine
 from openrouter.types import (
     BaseModel,
     Nullable,
@@ -136,6 +145,16 @@ OpenResponsesRequestToolUnion = Annotated[
 ]
 
 
+OpenResponsesRequestImageConfigTypedDict = TypeAliasType(
+    "OpenResponsesRequestImageConfigTypedDict", Union[str, float]
+)
+
+
+OpenResponsesRequestImageConfig = TypeAliasType(
+    "OpenResponsesRequestImageConfig", Union[str, float]
+)
+
+
 ServiceTier = Literal["auto",]
 
 
@@ -148,33 +167,57 @@ Truncation = Union[
 ]
 
 
-OrderTypedDict = TypeAliasType("OrderTypedDict", Union[ProviderName, str])
+OpenResponsesRequestOrderTypedDict = TypeAliasType(
+    "OpenResponsesRequestOrderTypedDict", Union[ProviderName, str]
+)
 
 
-Order = TypeAliasType(
-    "Order",
+OpenResponsesRequestOrder = TypeAliasType(
+    "OpenResponsesRequestOrder",
     Union[Annotated[ProviderName, PlainValidator(validate_open_enum(False))], str],
 )
 
 
-OnlyTypedDict = TypeAliasType("OnlyTypedDict", Union[ProviderName, str])
+OpenResponsesRequestOnlyTypedDict = TypeAliasType(
+    "OpenResponsesRequestOnlyTypedDict", Union[ProviderName, str]
+)
 
 
-Only = TypeAliasType(
-    "Only",
+OpenResponsesRequestOnly = TypeAliasType(
+    "OpenResponsesRequestOnly",
     Union[Annotated[ProviderName, PlainValidator(validate_open_enum(False))], str],
 )
 
 
-IgnoreTypedDict = TypeAliasType("IgnoreTypedDict", Union[ProviderName, str])
+OpenResponsesRequestIgnoreTypedDict = TypeAliasType(
+    "OpenResponsesRequestIgnoreTypedDict", Union[ProviderName, str]
+)
 
 
-Ignore = TypeAliasType(
-    "Ignore",
+OpenResponsesRequestIgnore = TypeAliasType(
+    "OpenResponsesRequestIgnore",
     Union[Annotated[ProviderName, PlainValidator(validate_open_enum(False))], str],
 )
 
 
+OpenResponsesRequestSortTypedDict = TypeAliasType(
+    "OpenResponsesRequestSortTypedDict",
+    Union[ProviderSortConfigTypedDict, ProviderSort, Any],
+)
+r"""The sorting strategy to use for this request, if \"order\" is not specified. When set, no load balancing is performed."""
+
+
+OpenResponsesRequestSort = TypeAliasType(
+    "OpenResponsesRequestSort",
+    Union[
+        ProviderSortConfig,
+        Annotated[ProviderSort, PlainValidator(validate_open_enum(False))],
+        Any,
+    ],
+)
+r"""The sorting strategy to use for this request, if \"order\" is not specified. When set, no load balancing is performed."""
+
+
 class OpenResponsesRequestMaxPriceTypedDict(TypedDict):
     r"""The object specifying the maximum price you want to pay for this request. USD price per million tokens, for prompt and completion."""
 
@@ -230,22 +273,22 @@ class OpenResponsesRequestProviderTypedDict(TypedDict):
     r"""Whether to restrict routing to only ZDR (Zero Data Retention) endpoints. When true, only endpoints that do not retain prompts will be used."""
     enforce_distillable_text: NotRequired[Nullable[bool]]
     r"""Whether to restrict routing to only models that allow text distillation. When true, only models where the author has allowed distillation will be used."""
-    order: NotRequired[Nullable[List[OrderTypedDict]]]
+    order: NotRequired[Nullable[List[OpenResponsesRequestOrderTypedDict]]]
     r"""An ordered list of provider slugs. The router will attempt to use the first provider in the subset of this list that supports your requested model, and fall back to the next if it is unavailable. If no providers are available, the request will fail with an error message."""
-    only: NotRequired[Nullable[List[OnlyTypedDict]]]
+    only: NotRequired[Nullable[List[OpenResponsesRequestOnlyTypedDict]]]
     r"""List of provider slugs to allow. If provided, this list is merged with your account-wide allowed provider settings for this request."""
-    ignore: NotRequired[Nullable[List[IgnoreTypedDict]]]
+    ignore: NotRequired[Nullable[List[OpenResponsesRequestIgnoreTypedDict]]]
     r"""List of provider slugs to ignore. If provided, this list is merged with your account-wide ignored provider settings for this request."""
     quantizations: NotRequired[Nullable[List[Quantization]]]
     r"""A list of quantization levels to filter the provider by."""
-    sort: NotRequired[Nullable[ProviderSort]]
+    sort: NotRequired[Nullable[OpenResponsesRequestSortTypedDict]]
     r"""The sorting strategy to use for this request, if \"order\" is not specified. When set, no load balancing is performed."""
     max_price: NotRequired[OpenResponsesRequestMaxPriceTypedDict]
     r"""The object specifying the maximum price you want to pay for this request. USD price per million tokens, for prompt and completion."""
-    min_throughput: NotRequired[Nullable[float]]
-    r"""The minimum throughput (in tokens per second) required for this request. Only providers serving the model with at least this throughput will be used."""
-    max_latency: NotRequired[Nullable[float]]
-    r"""The maximum latency (in seconds) allowed for this request. Only providers serving the model with better than this latency will be used."""
+    preferred_min_throughput: NotRequired[Nullable[PreferredMinThroughputTypedDict]]
+    r"""Preferred minimum throughput (in tokens per second). Can be a number (applies to p50) or an object with percentile-specific cutoffs. Endpoints below the threshold(s) may still be used, but are deprioritized in routing. When using fallback models, this may cause a fallback model to be used instead of the primary model if it meets the threshold."""
+    preferred_max_latency: NotRequired[Nullable[PreferredMaxLatencyTypedDict]]
+    r"""Preferred maximum latency (in seconds). Can be a number (applies to p50) or an object with percentile-specific cutoffs. Endpoints above the threshold(s) may still be used, but are deprioritized in routing. When using fallback models, this may cause a fallback model to be used instead of the primary model if it meets the threshold."""
 
 
 class OpenResponsesRequestProvider(BaseModel):
@@ -276,13 +319,13 @@ class OpenResponsesRequestProvider(BaseModel):
     enforce_distillable_text: OptionalNullable[bool] = UNSET
     r"""Whether to restrict routing to only models that allow text distillation. When true, only models where the author has allowed distillation will be used."""
 
-    order: OptionalNullable[List[Order]] = UNSET
+    order: OptionalNullable[List[OpenResponsesRequestOrder]] = UNSET
     r"""An ordered list of provider slugs. The router will attempt to use the first provider in the subset of this list that supports your requested model, and fall back to the next if it is unavailable. If no providers are available, the request will fail with an error message."""
 
-    only: OptionalNullable[List[Only]] = UNSET
+    only: OptionalNullable[List[OpenResponsesRequestOnly]] = UNSET
     r"""List of provider slugs to allow. If provided, this list is merged with your account-wide allowed provider settings for this request."""
 
-    ignore: OptionalNullable[List[Ignore]] = UNSET
+    ignore: OptionalNullable[List[OpenResponsesRequestIgnore]] = UNSET
     r"""List of provider slugs to ignore. If provided, this list is merged with your account-wide ignored provider settings for this request."""
 
     quantizations: OptionalNullable[
@@ -290,19 +333,17 @@ class OpenResponsesRequestProvider(BaseModel):
     ] = UNSET
     r"""A list of quantization levels to filter the provider by."""
 
-    sort: Annotated[
-        OptionalNullable[ProviderSort], PlainValidator(validate_open_enum(False))
-    ] = UNSET
+    sort: OptionalNullable[OpenResponsesRequestSort] = UNSET
     r"""The sorting strategy to use for this request, if \"order\" is not specified. When set, no load balancing is performed."""
 
     max_price: Optional[OpenResponsesRequestMaxPrice] = None
     r"""The object specifying the maximum price you want to pay for this request. USD price per million tokens, for prompt and completion."""
 
-    min_throughput: OptionalNullable[float] = UNSET
-    r"""The minimum throughput (in tokens per second) required for this request. Only providers serving the model with at least this throughput will be used."""
+    preferred_min_throughput: OptionalNullable[PreferredMinThroughput] = UNSET
+    r"""Preferred minimum throughput (in tokens per second). Can be a number (applies to p50) or an object with percentile-specific cutoffs. Endpoints below the threshold(s) may still be used, but are deprioritized in routing. When using fallback models, this may cause a fallback model to be used instead of the primary model if it meets the threshold."""
 
-    max_latency: OptionalNullable[float] = UNSET
-    r"""The maximum latency (in seconds) allowed for this request. Only providers serving the model with better than this latency will be used."""
+    preferred_max_latency: OptionalNullable[PreferredMaxLatency] = UNSET
+    r"""Preferred maximum latency (in seconds). Can be a number (applies to p50) or an object with percentile-specific cutoffs. Endpoints above the threshold(s) may still be used, but are deprioritized in routing. When using fallback models, this may cause a fallback model to be used instead of the primary model if it meets the threshold."""
 
     @model_serializer(mode="wrap")
     def serialize_model(self, handler):
@@ -318,8 +359,8 @@ class OpenResponsesRequestProvider(BaseModel):
             "quantizations",
             "sort",
             "max_price",
-            "min_throughput",
-            "max_latency",
+            "preferred_min_throughput",
+            "preferred_max_latency",
         ]
         nullable_fields = [
             "allow_fallbacks",
@@ -332,8 +373,8 @@ class OpenResponsesRequestProvider(BaseModel):
             "ignore",
             "quantizations",
             "sort",
-            "min_throughput",
-            "max_latency",
+            "preferred_min_throughput",
+            "preferred_max_latency",
         ]
         null_default_fields = []
 
@@ -381,33 +422,12 @@ class OpenResponsesRequestPluginResponseHealing(BaseModel):
 IDFileParser = Literal["file-parser",]
 
 
-OpenResponsesRequestPdfEngine = Union[
-    Literal[
-        "mistral-ocr",
-        "pdf-text",
-        "native",
-    ],
-    UnrecognizedStr,
-]
-
-
-class OpenResponsesRequestPdfTypedDict(TypedDict):
-    engine: NotRequired[OpenResponsesRequestPdfEngine]
-
-
-class OpenResponsesRequestPdf(BaseModel):
-    engine: Annotated[
-        Optional[OpenResponsesRequestPdfEngine],
-        PlainValidator(validate_open_enum(False)),
-    ] = None
-
-
 class OpenResponsesRequestPluginFileParserTypedDict(TypedDict):
     id: IDFileParser
     enabled: NotRequired[bool]
     r"""Set to false to disable the file-parser plugin for this request. Defaults to true."""
-    max_files: NotRequired[float]
-    pdf: NotRequired[OpenResponsesRequestPdfTypedDict]
+    pdf: NotRequired[PDFParserOptionsTypedDict]
+    r"""Options for PDF parsing."""
 
 
 class OpenResponsesRequestPluginFileParser(BaseModel):
@@ -416,30 +436,21 @@ class OpenResponsesRequestPluginFileParser(BaseModel):
     enabled: Optional[bool] = None
     r"""Set to false to disable the file-parser plugin for this request. Defaults to true."""
 
-    max_files: Optional[float] = None
-
-    pdf: Optional[OpenResponsesRequestPdf] = None
+    pdf: Optional[PDFParserOptions] = None
+    r"""Options for PDF parsing."""
 
 
 IDWeb = Literal["web",]
 
 
-OpenResponsesRequestEngine = Union[
-    Literal[
-        "native",
-        "exa",
-    ],
-    UnrecognizedStr,
-]
-
-
 class OpenResponsesRequestPluginWebTypedDict(TypedDict):
     id: IDWeb
     enabled: NotRequired[bool]
     r"""Set to false to disable the web-search plugin for this request. Defaults to true."""
     max_results: NotRequired[float]
     search_prompt: NotRequired[str]
-    engine: NotRequired[OpenResponsesRequestEngine]
+    engine: NotRequired[WebSearchEngine]
+    r"""The search engine to use for web search."""
 
 
 class OpenResponsesRequestPluginWeb(BaseModel):
@@ -453,8 +464,9 @@ class OpenResponsesRequestPluginWeb(BaseModel):
     search_prompt: Optional[str] = None
 
     engine: Annotated[
-        Optional[OpenResponsesRequestEngine], PlainValidator(validate_open_enum(False))
+        Optional[WebSearchEngine], PlainValidator(validate_open_enum(False))
     ] = None
+    r"""The search engine to use for web search."""
 
 
 IDModeration = Literal["moderation",]
@@ -468,11 +480,33 @@ class OpenResponsesRequestPluginModeration(BaseModel):
     id: IDModeration
 
 
+IDAutoRouter = Literal["auto-router",]
+
+
+class OpenResponsesRequestPluginAutoRouterTypedDict(TypedDict):
+    id: IDAutoRouter
+    enabled: NotRequired[bool]
+    r"""Set to false to disable the auto-router plugin for this request. Defaults to true."""
+    allowed_models: NotRequired[List[str]]
+    r"""List of model patterns to filter which models the auto-router can route between. Supports wildcards (e.g., \"anthropic/*\" matches all Anthropic models). When not specified, uses the default supported models list."""
+
+
+class OpenResponsesRequestPluginAutoRouter(BaseModel):
+    id: IDAutoRouter
+
+    enabled: Optional[bool] = None
+    r"""Set to false to disable the auto-router plugin for this request. Defaults to true."""
+
+    allowed_models: Optional[List[str]] = None
+    r"""List of model patterns to filter which models the auto-router can route between. Supports wildcards (e.g., \"anthropic/*\" matches all Anthropic models). When not specified, uses the default supported models list."""
+
+
 OpenResponsesRequestPluginUnionTypedDict = TypeAliasType(
     "OpenResponsesRequestPluginUnionTypedDict",
     Union[
         OpenResponsesRequestPluginModerationTypedDict,
         OpenResponsesRequestPluginResponseHealingTypedDict,
+        OpenResponsesRequestPluginAutoRouterTypedDict,
         OpenResponsesRequestPluginFileParserTypedDict,
         OpenResponsesRequestPluginWebTypedDict,
     ],
@@ -481,6 +515,7 @@ OpenResponsesRequestPluginUnionTypedDict = TypeAliasType(
 
 OpenResponsesRequestPluginUnion = Annotated[
     Union[
+        Annotated[OpenResponsesRequestPluginAutoRouter, Tag("auto-router")],
         Annotated[OpenResponsesRequestPluginModeration, Tag("moderation")],
         Annotated[OpenResponsesRequestPluginWeb, Tag("web")],
         Annotated[OpenResponsesRequestPluginFileParser, Tag("file-parser")],
@@ -490,16 +525,6 @@ OpenResponsesRequestPluginUnion = Annotated[
 ]
 
 
-OpenResponsesRequestRoute = Union[
-    Literal[
-        "fallback",
-        "sort",
-    ],
-    UnrecognizedStr,
-]
-r"""Routing strategy for multiple models: \"fallback\" (default) uses secondary models as backups, \"sort\" sorts all endpoints together by routing criteria."""
-
-
 class OpenResponsesRequestTypedDict(TypedDict):
     r"""Request schema for Responses endpoint"""
 
@@ -520,7 +545,15 @@ class OpenResponsesRequestTypedDict(TypedDict):
     max_output_tokens: NotRequired[Nullable[float]]
     temperature: NotRequired[Nullable[float]]
     top_p: NotRequired[Nullable[float]]
+    top_logprobs: NotRequired[Nullable[int]]
+    max_tool_calls: NotRequired[Nullable[int]]
+    presence_penalty: NotRequired[Nullable[float]]
+    frequency_penalty: NotRequired[Nullable[float]]
     top_k: NotRequired[float]
+    image_config: NotRequired[Dict[str, OpenResponsesRequestImageConfigTypedDict]]
+    r"""Provider-specific image configuration options. Keys and values vary by model/provider. See https://openrouter.ai/docs/features/multimodal/image-generation for more details."""
+    modalities: NotRequired[List[ResponsesOutputModality]]
+    r"""Output modalities for the response. Supported values are \"text\" and \"image\"."""
     prompt_cache_key: NotRequired[Nullable[str]]
     previous_response_id: NotRequired[Nullable[str]]
     prompt: NotRequired[Nullable[OpenAIResponsesPromptTypedDict]]
@@ -535,8 +568,6 @@ class OpenResponsesRequestTypedDict(TypedDict):
     r"""When multiple model providers are available, optionally indicate your routing preference."""
     plugins: NotRequired[List[OpenResponsesRequestPluginUnionTypedDict]]
     r"""Plugins you want to enable for this request, including their settings."""
-    route: NotRequired[Nullable[OpenResponsesRequestRoute]]
-    r"""Routing strategy for multiple models: \"fallback\" (default) uses secondary models as backups, \"sort\" sorts all endpoints together by routing criteria."""
     user: NotRequired[str]
     r"""A unique identifier representing your end-user, which helps distinguish between different users of your app. This allows your app to identify specific users in case of abuse reports, preventing your entire app from being affected by the actions of individual users. Maximum of 128 characters."""
     session_id: NotRequired[str]
@@ -576,8 +607,28 @@ class OpenResponsesRequest(BaseModel):
 
     top_p: OptionalNullable[float] = UNSET
 
+    top_logprobs: OptionalNullable[int] = UNSET
+
+    max_tool_calls: OptionalNullable[int] = UNSET
+
+    presence_penalty: OptionalNullable[float] = UNSET
+
+    frequency_penalty: OptionalNullable[float] = UNSET
+
     top_k: Optional[float] = None
 
+    image_config: Optional[Dict[str, OpenResponsesRequestImageConfig]] = None
+    r"""Provider-specific image configuration options. Keys and values vary by model/provider. See https://openrouter.ai/docs/features/multimodal/image-generation for more details."""
+
+    modalities: Optional[
+        List[
+            Annotated[
+                ResponsesOutputModality, PlainValidator(validate_open_enum(False))
+            ]
+        ]
+    ] = None
+    r"""Output modalities for the response. Supported values are \"text\" and \"image\"."""
+
     prompt_cache_key: OptionalNullable[str] = UNSET
 
     previous_response_id: OptionalNullable[str] = UNSET
@@ -615,12 +666,6 @@ class OpenResponsesRequest(BaseModel):
     plugins: Optional[List[OpenResponsesRequestPluginUnion]] = None
     r"""Plugins you want to enable for this request, including their settings."""
 
-    route: Annotated[
-        OptionalNullable[OpenResponsesRequestRoute],
-        PlainValidator(validate_open_enum(False)),
-    ] = UNSET
-    r"""Routing strategy for multiple models: \"fallback\" (default) uses secondary models as backups, \"sort\" sorts all endpoints together by routing criteria."""
-
     user: Optional[str] = None
     r"""A unique identifier representing your end-user, which helps distinguish between different users of your app. This allows your app to identify specific users in case of abuse reports, preventing your entire app from being affected by the actions of individual users. Maximum of 128 characters."""
 
@@ -643,7 +688,13 @@ class OpenResponsesRequest(BaseModel):
             "max_output_tokens",
             "temperature",
             "top_p",
+            "top_logprobs",
+            "max_tool_calls",
+            "presence_penalty",
+            "frequency_penalty",
             "top_k",
+            "image_config",
+            "modalities",
             "prompt_cache_key",
             "previous_response_id",
             "prompt",
@@ -656,7 +707,6 @@ class OpenResponsesRequest(BaseModel):
             "stream",
             "provider",
             "plugins",
-            "route",
             "user",
             "session_id",
         ]
@@ -668,6 +718,10 @@ class OpenResponsesRequest(BaseModel):
             "max_output_tokens",
             "temperature",
             "top_p",
+            "top_logprobs",
+            "max_tool_calls",
+            "presence_penalty",
+            "frequency_penalty",
             "prompt_cache_key",
             "previous_response_id",
             "prompt",
@@ -676,7 +730,6 @@ class OpenResponsesRequest(BaseModel):
             "safety_identifier",
             "truncation",
             "provider",
-            "route",
         ]
         null_default_fields = []
 

openrouter/components/parameter.py

@@ -28,6 +28,7 @@ Parameter = Union[
         "parallel_tool_calls",
         "include_reasoning",
         "reasoning",
+        "reasoning_effort",
         "web_search_options",
         "verbosity",
     ],

openrouter/components/pdfparserengine.py

@@ -0,0 +1,16 @@
+"""Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT."""
+
+from __future__ import annotations
+from openrouter.types import UnrecognizedStr
+from typing import Literal, Union
+
+
+PDFParserEngine = Union[
+    Literal[
+        "mistral-ocr",
+        "pdf-text",
+        "native",
+    ],
+    UnrecognizedStr,
+]
+r"""The engine to use for parsing PDF files."""

openrouter/components/pdfparseroptions.py

@@ -0,0 +1,25 @@
+"""Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT."""
+
+from __future__ import annotations
+from .pdfparserengine import PDFParserEngine
+from openrouter.types import BaseModel
+from openrouter.utils import validate_open_enum
+from pydantic.functional_validators import PlainValidator
+from typing import Optional
+from typing_extensions import Annotated, NotRequired, TypedDict
+
+
+class PDFParserOptionsTypedDict(TypedDict):
+    r"""Options for PDF parsing."""
+
+    engine: NotRequired[PDFParserEngine]
+    r"""The engine to use for parsing PDF files."""
+
+
+class PDFParserOptions(BaseModel):
+    r"""Options for PDF parsing."""
+
+    engine: Annotated[
+        Optional[PDFParserEngine], PlainValidator(validate_open_enum(False))
+    ] = None
+    r"""The engine to use for parsing PDF files."""

openrouter/components/percentilelatencycutoffs.py

@@ -0,0 +1,71 @@
+"""Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT."""
+
+from __future__ import annotations
+from openrouter.types import (
+    BaseModel,
+    Nullable,
+    OptionalNullable,
+    UNSET,
+    UNSET_SENTINEL,
+)
+from pydantic import model_serializer
+from typing_extensions import NotRequired, TypedDict
+
+
+class PercentileLatencyCutoffsTypedDict(TypedDict):
+    r"""Percentile-based latency cutoffs. All specified cutoffs must be met for an endpoint to be preferred."""
+
+    p50: NotRequired[Nullable[float]]
+    r"""Maximum p50 latency (seconds)"""
+    p75: NotRequired[Nullable[float]]
+    r"""Maximum p75 latency (seconds)"""
+    p90: NotRequired[Nullable[float]]
+    r"""Maximum p90 latency (seconds)"""
+    p99: NotRequired[Nullable[float]]
+    r"""Maximum p99 latency (seconds)"""
+
+
+class PercentileLatencyCutoffs(BaseModel):
+    r"""Percentile-based latency cutoffs. All specified cutoffs must be met for an endpoint to be preferred."""
+
+    p50: OptionalNullable[float] = UNSET
+    r"""Maximum p50 latency (seconds)"""
+
+    p75: OptionalNullable[float] = UNSET
+    r"""Maximum p75 latency (seconds)"""
+
+    p90: OptionalNullable[float] = UNSET
+    r"""Maximum p90 latency (seconds)"""
+
+    p99: OptionalNullable[float] = UNSET
+    r"""Maximum p99 latency (seconds)"""
+
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = ["p50", "p75", "p90", "p99"]
+        nullable_fields = ["p50", "p75", "p90", "p99"]
+        null_default_fields = []
+
+        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
+
+        return m

openrouter/components/percentilestats.py

@@ -0,0 +1,34 @@
+"""Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT."""
+
+from __future__ import annotations
+from openrouter.types import BaseModel
+from typing_extensions import TypedDict
+
+
+class PercentileStatsTypedDict(TypedDict):
+    r"""Latency percentiles in milliseconds over the last 30 minutes. Latency measures time to first token. Only visible when authenticated with an API key or cookie; returns null for unauthenticated requests."""
+
+    p50: float
+    r"""Median (50th percentile)"""
+    p75: float
+    r"""75th percentile"""
+    p90: float
+    r"""90th percentile"""
+    p99: float
+    r"""99th percentile"""
+
+
+class PercentileStats(BaseModel):
+    r"""Latency percentiles in milliseconds over the last 30 minutes. Latency measures time to first token. Only visible when authenticated with an API key or cookie; returns null for unauthenticated requests."""
+
+    p50: float
+    r"""Median (50th percentile)"""
+
+    p75: float
+    r"""75th percentile"""
+
+    p90: float
+    r"""90th percentile"""
+
+    p99: float
+    r"""99th percentile"""

openrouter/components/percentilethroughputcutoffs.py

@@ -0,0 +1,71 @@
+"""Code generated by Speakeasy (https://speakeasy.com). DO NOT EDIT."""
+
+from __future__ import annotations
+from openrouter.types import (
+    BaseModel,
+    Nullable,
+    OptionalNullable,
+    UNSET,
+    UNSET_SENTINEL,
+)
+from pydantic import model_serializer
+from typing_extensions import NotRequired, TypedDict
+
+
+class PercentileThroughputCutoffsTypedDict(TypedDict):
+    r"""Percentile-based throughput cutoffs. All specified cutoffs must be met for an endpoint to be preferred."""
+
+    p50: NotRequired[Nullable[float]]
+    r"""Minimum p50 throughput (tokens/sec)"""
+    p75: NotRequired[Nullable[float]]
+    r"""Minimum p75 throughput (tokens/sec)"""
+    p90: NotRequired[Nullable[float]]
+    r"""Minimum p90 throughput (tokens/sec)"""
+    p99: NotRequired[Nullable[float]]
+    r"""Minimum p99 throughput (tokens/sec)"""
+
+
+class PercentileThroughputCutoffs(BaseModel):
+    r"""Percentile-based throughput cutoffs. All specified cutoffs must be met for an endpoint to be preferred."""
+
+    p50: OptionalNullable[float] = UNSET
+    r"""Minimum p50 throughput (tokens/sec)"""
+
+    p75: OptionalNullable[float] = UNSET
+    r"""Minimum p75 throughput (tokens/sec)"""
+
+    p90: OptionalNullable[float] = UNSET
+    r"""Minimum p90 throughput (tokens/sec)"""
+
+    p99: OptionalNullable[float] = UNSET
+    r"""Minimum p99 throughput (tokens/sec)"""
+
+    @model_serializer(mode="wrap")
+    def serialize_model(self, handler):
+        optional_fields = ["p50", "p75", "p90", "p99"]
+        nullable_fields = ["p50", "p75", "p90", "p99"]
+        null_default_fields = []
+
+        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
+
+        return m