How to stream chat model responses

All chat models implement the Runnable interface, which comes with a default implementations of standard runnable methods (i.e. ainvoke, batch, abatch, stream, astream, astream_events).

The default streaming implementation provides anIterator (or AsyncIterator for asynchronous streaming) that yields a single value: the final output from the underlying chat model provider.

tip

The default implementation does not provide support for token-by-token streaming, but it ensures that the the model can be swapped in for any other model as it supports the same standard interface.

The ability to stream the output token-by-token depends on whether the provider has implemented proper streaming support.

See which integrations support token-by-token streaming here.

Sync streaming

Below we use a | to help visualize the delimiter between tokens.

from langchain_anthropic.chat_models import ChatAnthropic

chat = ChatAnthropic(model="claude-3-haiku-20240307")
for chunk in chat.stream("Write me a 1 verse song about goldfish on the moon"):
    print(chunk.content, end="|", flush=True)

API Reference:

• ChatAnthropic

Here| is| a| |1| |verse| song| about| gol|dfish| on| the| moon|:|

Floating| up| in| the| star|ry| night|,|
Fins| a|-|gl|im|mer| in| the| pale| moon|light|.|
Gol|dfish| swimming|,| peaceful| an|d free|,|
Se|ren|ely| |drif|ting| across| the| lunar| sea|.|

Async Streaming

from langchain_anthropic.chat_models import ChatAnthropic

chat = ChatAnthropic(model="claude-3-haiku-20240307")
async for chunk in chat.astream("Write me a 1 verse song about goldfish on the moon"):
    print(chunk.content, end="|", flush=True)

API Reference:

• ChatAnthropic

Here| is| a| |1| |verse| song| about| gol|dfish| on| the| moon|:|

Floating| up| above| the| Earth|,|
Gol|dfish| swim| in| alien| m|irth|.|
In| their| bowl| of| lunar| dust|,|
Gl|it|tering| scales| reflect| the| trust|
Of| swimming| free| in| this| new| worl|d,|
Where| their| aqu|atic| dream|'s| unf|ur|le|d.|

Astream events

Chat models also support the standard astream events method.

This method is useful if you're streaming output from a larger LLM application that contains multiple steps (e.g., an LLM chain composed of a prompt, llm and parser).

from langchain_anthropic.chat_models import ChatAnthropic

chat = ChatAnthropic(model="claude-3-haiku-20240307")
idx = 0

async for event in chat.astream_events(
    "Write me a 1 verse song about goldfish on the moon", version="v1"
):
    idx += 1
    if idx >= 5:  # Truncate the output
        print("...Truncated")
        break
    print(event)

API Reference:

• ChatAnthropic

{'event': 'on_chat_model_start', 'run_id': '08da631a-12a0-4f07-baee-fc9a175ad4ba', 'name': 'ChatAnthropic', 'tags': [], 'metadata': {}, 'data': {'input': 'Write me a 1 verse song about goldfish on the moon'}}
{'event': 'on_chat_model_stream', 'run_id': '08da631a-12a0-4f07-baee-fc9a175ad4ba', 'tags': [], 'metadata': {}, 'name': 'ChatAnthropic', 'data': {'chunk': AIMessageChunk(content='Here', id='run-08da631a-12a0-4f07-baee-fc9a175ad4ba')}}
{'event': 'on_chat_model_stream', 'run_id': '08da631a-12a0-4f07-baee-fc9a175ad4ba', 'tags': [], 'metadata': {}, 'name': 'ChatAnthropic', 'data': {'chunk': AIMessageChunk(content="'s", id='run-08da631a-12a0-4f07-baee-fc9a175ad4ba')}}
{'event': 'on_chat_model_stream', 'run_id': '08da631a-12a0-4f07-baee-fc9a175ad4ba', 'tags': [], 'metadata': {}, 'name': 'ChatAnthropic', 'data': {'chunk': AIMessageChunk(content=' a', id='run-08da631a-12a0-4f07-baee-fc9a175ad4ba')}}
...Truncated