OllamaClient.chatStream

Streaming chat — calls onChunk for every assistant token.

Each chunk contains "message": {"role": "assistant", "content": "<token>"}. The final chunk has "done": true and carries usage metadata.

class OllamaClient

@safe

void

chatStream

(

string model

,

Message[] messages

,

StreamCallback onChunk

,

Tool[] tools = null

,

JSONValue format = JSONValue.init

,

string keepAlive = null

,

OllamaOptions opts = OllamaOptions.init

)

Parameters

model string: Model name.
messages Message[]: Conversation history.
onChunk StreamCallback: Callback invoked per chunk; must be @safe.
tools Tool[]: Optional tool definitions.
format JSONValue: Structured output schema or JSONValue("json").
keepAlive string: How long to keep the model loaded.
opts OllamaOptions: Typed generation options.

Meta

Source

See Implementation