18 lines
466 B
Python
18 lines
466 B
Python
from llama_index.core.llms.types import (
|
|
ChatMessage,
|
|
ChatResponse,
|
|
ChatResponseGen,
|
|
MessageRole,
|
|
)
|
|
from llama_index.types import TokenGen
|
|
|
|
|
|
def response_gen_from_query_engine(response_gen: TokenGen) -> ChatResponseGen:
|
|
response_str = ""
|
|
for token in response_gen:
|
|
response_str += token
|
|
yield ChatResponse(
|
|
message=ChatMessage(role=MessageRole.ASSISTANT, content=response_str),
|
|
delta=token,
|
|
)
|