types library
Classes
- ChatCompletion
-
A complete Chat Completion response (non-streaming). Mirrors OpenAI
chat.completion. - ChatCompletionChoice
- ChatCompletionChunk
-
A single Chat Completion chunk (streaming). Mirrors OpenAI
chat.completion.chunk. - ChatCompletionChunkChoice
- ChatCompletionChunkFunction
- ChatCompletionChunkToolCall
- A tool call delta in a ChatCompletionChunk.
- ChatCompletionDelta
- ChatCompletionFunction
- ChatCompletionMessage
- ChatCompletionToolCall
- A tool call in a ChatCompletionMessage.
- ChatCompletionUsage
- Token usage statistics (shared by streaming and non-streaming).
- CompletionTokensDetails
- ContentPart
- ContentPartFile
- ContentPartFileBase64
- ContentPartFileReference
- ContentPartFileUrl
- ContentPartImage
- ContentPartReasoning
- ContentPartText
- ContentPartToolCall
- ContentPartToolResult
- ContentPartUnknown
- FileBytes
- FileBytesBase64
- FileBytesBinary
- FileData
- FileDataData
- FileDataReference
- FileDataText
- FileDataUrl
- FinishReason
-
Why generation stopped. Mirrors
FinishReason.ts. - FunctionTool
-
A function tool definition. Mirrors
FunctionTool.ts. - GenerateContent
-
A content item in a
GenerateResult. MirrorsGenerateContent.ts(aimux-core/src/result.rs). - GenerateContentFile
-
A file generated by the model (e.g. an image or document). The
datafield is aFileDatatagged union; there is nofilenamefield. - GenerateContentReasoning
- A reasoning / thinking segment produced by the model.
- GenerateContentSource
- A source / citation (e.g. URL citation from search-preview models).
- GenerateContentText
- Generated text.
- GenerateContentToolCall
- A tool call requested by the model.
- GenerateContentToolResult
-
A tool result from a provider-executed tool (e.g. xAI file_search,
web_search). Emitted alongside the preceding
ToolCall. - GenerateContentUnknown
-
Fallback for unknown/forward-compatible
GenerateContentvariants. - GenerateObjectResult
-
Result of
generate_object(user-facing, M12). The parsed JSON object plus convenience fields from the underlyinggenerate_textcall. - GenerateResult
- GenerateTextOptions
-
User-facing options for
generate_text/stream_text. MirrorsGenerateTextOptions.ts. - GenerateTextResult
-
Result of
generate_text(user-facing). MirrorsGenerateTextResult.ts. - ModelMessage
-
A single user-facing chat message. Mirrors
ModelMessage.ts. - PromptTokensDetails
- ResponseMetadata
- StreamPart
-
A single chunk in the stream returned by
stream_text. MirrorsStreamPart.ts(aimux-core/src/stream_part.rs). - StreamPartError
- An error occurred mid-stream.
- StreamPartFile
-
A file generated by the model (e.g. an image or document). The
datafield is aFileDatatagged union; there is nofilenamefield. - StreamPartFinish
- Final chunk — carries usage, finish reason, and metadata.
- StreamPartRaw
-
A raw chunk from the provider (for debugging, when
include_raw_chunksis set). - StreamPartReasoningDelta
- A delta of a reasoning / thinking segment.
- StreamPartReasoningEnd
- End of a reasoning / thinking segment.
- StreamPartReasoningStart
- Start of a reasoning / thinking segment.
- StreamPartResponseMetadata
- Response metadata (id, timestamp, model_id).
- StreamPartSource
- A source / citation (e.g. URL citation from search-preview models).
- StreamPartStreamStart
- First chunk — carries warnings from the provider.
- StreamPartTextDelta
- A delta of generated text.
- StreamPartTextEnd
- End of a text segment.
- StreamPartTextStart
- Start of a text segment.
- StreamPartToolCall
- A complete tool call (alternative to the start/delta/end flow).
- StreamPartToolInputDelta
- A delta of tool call input (partial JSON).
- StreamPartToolInputEnd
- End of a tool call's input streaming.
- StreamPartToolInputStart
- Start of a tool call's input streaming.
- StreamPartToolResult
- A tool result (provider-executed tools).
- StreamPartUnknown
-
Fallback for unknown/forward-compatible
StreamPartvariants. - StreamTextResultAggregated
-
Aggregated result of
stream_text().consume()(M11). MirrorsGenerateTextResult's user-facing fields (withoutraw, since streaming has noGenerateResultequivalent). - TimeoutConfiguration
-
Per-call timeout configuration. Mirrors
TimeoutConfiguration.ts. - TokenUsage
-
Token usage detail (with cache breakdown). Mirrors
TokenUsage.ts. - Tool
-
A tool definition: function or provider tool. Mirrors
Tool.ts. - ToolCall
-
A tool call requested by the model. Mirrors
ToolCall.ts. - ToolChoice
-
How the model should choose tools. Mirrors
ToolChoice.ts. - Usage
-
Token usage statistics. Mirrors
Usage.ts.