openai.internal.chat_invoke_stream
The streaming entry point: the same request as `chat_invoke` with `stream: true`, wrapped in an `ai.stream.TurnStream` over the SSE connection.
Reference version
Signature
function openai.internal.chat_invoke_stream(
compat: openai.internal.ChatCompat,
params: openai.internal.ChatParams,
input: ai.ModelTurnInput,
capture_wire: bool = …,
request_timeout_ms: int | null = …,
time_to_first_token_timeout_ms: int | null = …
) -> ai.stream.TurnStream throws baml.errors.InvalidArgument | baml.errors.Io | baml.errors.Timeout | baml.json.SerializationError | ai.errors.Failure
The streaming entry point: the same request as chat_invoke with
stream: true, wrapped in an ai.stream.TurnStream over the SSE connection.
Source:<builtin>/openai/ns_internal/chat.bamlbytes 63568–66047