OllamaClient.generateStream

Streaming text generation — calls onChunk for every response token.

Each call to onChunk receives one NDJSON chunk. The chunk contains a "response" string token and a boolean "done". The final chunk has "done": true and carries usage/timing metadata.

class OllamaClient

@safe

void

generateStream

(

string model

string prompt

StreamCallback onChunk

string system = null

string[] images = null

JSONValue format = JSONValue.init

string keepAlive = null

OllamaOptions opts = OllamaOptions.init

)

Parameters

model string: Model name (e.g. "llama3.1:8b").
prompt string: Input prompt.
onChunk StreamCallback: Callback invoked per chunk; must be @safe.
system string: Optional system prompt.
images string[]: Optional base64-encoded images (multimodal).
format JSONValue: Structured output: JSONValue("json") or JSON Schema.
keepAlive string: How long to keep the model loaded.
opts OllamaOptions: Typed generation options.

OllamaClient.generateStream

Parameters

Meta

Source