langserve

langserve copied to clipboard

Setting max_execution_time of Agent does not work for /stream

1

Based on https://python.langchain.com/docs/modules/agents/how_to/max_time_limit we can set a timeout for agents and a message will be returned to the user, this works in LangChain. However when using this feature with LangServe...

fatimatayeb

Bug: AgentExecutor configurable does not propagate configuration from agent.

1

Users reasonably expect that when an agent is configurable, then the configuration information is properly propagated via the Agent Executor. However, this is not the case at the moment. Bug...

eyurtsev

🔒 Easier Auth -- I'd like to pass a fastapi dependency to the add_routes method to secure them

5

- right now I have to add my auth on langserve through middleware - this works in prod, so that's great! - in dev, it makes it so I have...

randypitcherii

🔒 Add auth support to playground

2

Add support to playground for auth

eyurtsev

May be difficult to distinguish chat history from each users when wrap conversional agent by langserve

6

ConversationBufferMemory is useful in conversational agents, like codes below ``` memory = ConversationBufferMemory(memory_key="chat_history", return_messages=True) chain = initialize_agent(tools, llm, agent=AgentType.CHAT_CONVERSATIONAL_REACT_DESCRIPTION, verbose=True, memory=memory, ) ``` However, when integrated with langserve, it may...

ChengyangDu

Streaming RemoteRunnable not working w/ rag-chroma-private

2

Testing [rag-chroma-private](https://github.com/langchain-ai/langchain/tree/master/templates/rag-chroma-private). for - ``` from langserve.client import RemoteRunnable rag_app = RemoteRunnable('http://0.0.0.0:8001/rag_chroma_private/') ``` this returns a generator - ``` rag_app.stream("How does agent memory work?") ``` and this returns the answer...

rlancemartin

How to do Human in Loop in /stream Endpoint

1

LangServe can stream output using `/stream` Endpoint. However, if I want to send an event and wait for human feedback, wait for a parameter value or confirm, for example. It...

wolvever

Need help to migrate my custom agent to LCEL to use with langserve

6

Hi, team-langchain, I have an agent that uses memory, user-authentication as well as function calling. I'd like to migrate it to langserve in production but couldn't find anything as complex...

pokidyshev

Add option to disable endpoints to only expose a subset

1

eyurtsev

Streaming doesn't seem to work properly when using an AzureChatOpenAI model.

5

Hello, I'm unable to get an SSE streaming endpoint working correctly when using an Azure-hosted OpenAI model (using the `AzureChatOpenAI` class). I'm using a simple LCEL chain: `chain = promptTemplate...

mlamothe-zz