WithM sets the M parameter.

WithMaxCompletionTokens is used to set the max completion tokens for the request.

WithMaxCompletionTokens is used to set the max completion tokens for the request.

WithMaxToolCalls sets the maximum number of tool-calling rounds. This option is only supported for the ResponsesAPIChatModel. The value must be in the

WithMetadata attaches a set of up to 16 key-value pairs to an object, which can be useful for storing additional information in a structured format. T

WithModels provider an array of model IDs in priority order. If the first model returns an error, OpenRouter will automatically try the next model in

WithNList sets the NList parameter.

WithNList sets the NList parameter.

WithNList sets the NList parameter.

WithNList sets the NList parameter.

WithNList sets the number of cluster units.

WithPartition returns an option that sets the target partition for insertion.

Deprecated: use WithCache instead. WithPrefixCache creates an option to specify a context ID for the request. The context ID is typically obtained fro

WithReasoning provider advanced reasoning capabilities, allowing models to show their internal reasoning process with configurable effort、summary fiel

WithRequestPayloadModifier registers a payload modifier to customize the serialized request based on input messages. This is useful for OpenAI-compati

WithResourceDetector configures resource detector

WithResponseChunkMessageModifier registers a message modifier to transform the output message chunk using the raw response body. When end is true, msg

WithResponseFormat sets the response format for the request. When set, it overrides the ResponseFormat field configured in Config for this specific re

WithResponseFormat sets the response format for structured JSON output.

WithResponseMessageModifier registers a message modifier to transform the output message using the raw response body. This is useful for extracting pr

WithResponsesTruncation sets how to handle context that exceeds the model's context window. Available only for Responses API.

WithSparseVector sets the sparse vector for the retrieve query. For example, a stored vector {"feature_0": 0.12, "feature_1": 1.2, "feature_2": 3.0}.

WithSparseVector sets the sparse vector for the retrieve query. For example, a stored vector {"feature_0": 0.12, "feature_1": 1.2, "feature_2": 3.0}.

defaultIDGenerator keeps the original ID

defaultIDGenerator keeps the original ID

defaultIDGenerator keeps the original ID

defaultIDGenerator keeps the original ID

getMsgExtraInt64Value extracts an integer value from message extra, trying T first, then falling back to float64/int64/int to handle JSON unmarshal ty