Skip to main content Module common Copy item path Source AudioUrl ChatLogProbsContent CompletionTokensDetails ContextManagementEntry A single entry in the Responses API context_management array. ErrorDetail ErrorResponse Function FunctionCallDelta FunctionCallResponse FunctionChoice Function choice specification for ToolChoice::Function ImageUrl InputAudio JsonSchemaFormat LogProbs PromptTokenUsageInfo Redacted ResponsePrompt Reference to a prompt template and its variables. StreamOptions StringOrArrayIter Iterator over StringOrArray that yields string references without cloning Tool ToolCall ToolCallDelta TopLogProb Usage UsageInfo Usage information (used by rerank and other endpoints) VideoUrl ChatLogProbs ContentPart ContextManagementType Type tag for ContextManagementEntrycompaction is
defined by the spec; the enum is kept small so unknown values serde-fail
(consistent with P5’s fail-fast direction). ConversationRef Reference to a conversation the response belongs to. Detail Image detail level for PromptVariableTyped::ResponseInputImagecrate::responses::ResponseContentPart::InputImage"low" | "high" | "auto" | "original". FunctionCall Deprecated function_call field from the OpenAI API.
Can be "none", "auto", or {"name": "function_name"}. InputIds LoRAPath LoRA adapter path - can be single path or batch of paths (SGLang extension) PromptCacheRetention Retention policy for prompt-cache entries on the Responses API. PromptVariable A prompt variable value: plain string or a typed input (text, image, file). PromptVariableTyped Typed prompt variable input. ResponseFormat StringOrArray A type that can be either a single string or an array of strings ToolChoice Tool choice for both Chat Completion and Responses APIs ToolChoiceValue Tool choice value for simple string options ToolReference Tool reference for ToolChoice::AllowedTools GenerationRequest Trait for unified access to generation request properties
Implemented by ChatCompletionRequest, CompletionRequest, GenerateRequest,
EmbeddingRequest, RerankRequest, and ResponsesRequest default_true Helper function for serde default value (returns true) default_unknown_model Default model for endpoints where model is optional (e.g., /generate).
Uses UNKNOWN_MODEL_ID so routers treat it as “any available worker.” deserialize_null_as_false Deserialize a bool that also accepts JSON null (mapped to false). is_false Helper for #[serde(skip_serializing_if = "is_false")] on default-false flags. is_true Helper for #[serde(skip_serializing_if = "is_true")] on default-true flags. validate_stop Validates stop sequences (max 4, non-empty strings)
Used by both ChatCompletionRequest and ResponsesRequest