Leak (?) when using object member in a RunnableSequence

[maxfriedrich]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

import gc

from langchain_core.runnables import (
    RunnableLambda,
    RunnablePassthrough,
)


def log_memory_usage():
    gc.collect()
    objs = gc.get_objects()
    print(f"Total objects: {len(objs)}")


class ThingWithRunnable:
    def __init__(self, func):
        self.func = func
        self.runnable_lambda = RunnableLambda(func)
        self.expensive = {i: [i] for i in range(1000000)}

    def call(self, inputs):
        return self.runnable_lambda.invoke(inputs)

    def __del__(self):
        print(">>>> Cleaning up")


def func(inp):
    return {"result": "ok"}


def works():
    thing = ThingWithRunnable(func)
    thing.call({})


def works2():
    thing = ThingWithRunnable(func)
    (RunnableLambda(thing.func) | RunnablePassthrough()).invoke({})


def broken():
    thing = ThingWithRunnable(func)
    (thing.call | RunnablePassthrough()).invoke({})


def broken2():
    thing = ThingWithRunnable(func)
    (RunnableLambda(thing.call) | RunnablePassthrough()).invoke({})


if __name__ == "__main__":
    gc.freeze()
    print("Works:")
    for _ in range(2):
        works()
        log_memory_usage()

    print("Works 2:")
    for _ in range(2):
        works2()
        log_memory_usage()

    print("Broken:")
    for _ in range(2):
        broken()
        log_memory_usage()

    print("Broken 2:")
    for _ in range(2):
        broken2()
        log_memory_usage()

Description

Hi, we encountered a memory leak in our app and simplified it down to this. When creating a new RunnableSequence where a member of an object (thing.call) that invokes a Runnable is followed by something else like a RunnablePassthrough, the memory is not cleaned up.

There seems to be a reference to thing still held somewhere which is why the GC doesn't remove the object from memory.

Output:

Works:
>>>> Cleaning up
Total objects: 24630
>>>> Cleaning up
Total objects: 24624
Works 2:
>>>> Cleaning up
Total objects: 24904
>>>> Cleaning up
Total objects: 24904
Works 3:
>>>> Cleaning up
Total objects: 24965
>>>> Cleaning up
Total objects: 24965
Broken:
Total objects: 1024972
Total objects: 2024978
Broken 2:
Total objects: 3024984
Total objects: 4024990

In our real app, we were able to work around it by creating two Runnables, the first with the object member and the second with further steps, and invoking them one after the other.

Is this expected and we're using it wrong or is it a bug? Thank you!

System Info

System Information

OS: Darwin
OS Version: Darwin Kernel Version 24.3.0: Thu Jan 2 20:24:24 PST 2025; root:xnu-11215.81.4~3/RELEASE_ARM64_T6030
Python Version: 3.11.10 (main, Oct 16 2024, 08:56:36) [Clang 18.1.8 ]

Package Information

langchain_core: 0.3.48
langchain: 0.3.21
langsmith: 0.3.11
langchain_aws: 0.2.16
langchain_openai: 0.3.9
langchain_text_splitters: 0.3.7
langgraph_sdk: 0.1.58

Optional packages not installed

langserve

Other Dependencies

async-timeout<5.0.0,>=4.0.0;: Installed. No version info available.
boto3: 1.37.16
httpx: 0.28.1
jsonpatch<2.0,>=1.33: Installed. No version info available.
langchain-anthropic;: Installed. No version info available.
langchain-aws;: Installed. No version info available.
langchain-azure-ai;: Installed. No version info available.
langchain-cohere;: Installed. No version info available.
langchain-community;: Installed. No version info available.
langchain-core<1.0.0,>=0.3.45: Installed. No version info available.
langchain-deepseek;: Installed. No version info available.
langchain-fireworks;: Installed. No version info available.
langchain-google-genai;: Installed. No version info available.
langchain-google-vertexai;: Installed. No version info available.
langchain-groq;: Installed. No version info available.
langchain-huggingface;: Installed. No version info available.
langchain-mistralai;: Installed. No version info available.
langchain-ollama;: Installed. No version info available.
langchain-openai;: Installed. No version info available.
langchain-text-splitters<1.0.0,>=0.3.7: Installed. No version info available.
langchain-together;: Installed. No version info available.
langchain-xai;: Installed. No version info available.
langsmith-pyo3: Installed. No version info available.
langsmith<0.4,>=0.1.125: Installed. No version info available.
langsmith<0.4,>=0.1.17: Installed. No version info available.
numpy: 1.26.4
openai<2.0.0,>=1.66.3: Installed. No version info available.
orjson: 3.10.15
packaging: 24.2
packaging<25,>=23.2: Installed. No version info available.
pydantic: 2.10.5
pydantic<3.0.0,>=2.5.2;: Installed. No version info available.
pydantic<3.0.0,>=2.7.4: Installed. No version info available.
pydantic<3.0.0,>=2.7.4;: Installed. No version info available.
pytest: 8.3.4
PyYAML>=5.3: Installed. No version info available.
requests: 2.32.3
requests-toolbelt: 1.0.0
requests<3,>=2: Installed. No version info available.
rich: 13.9.4
SQLAlchemy<3,>=1.4: Installed. No version info available.
tenacity!=8.4.0,<10.0.0,>=8.1.0: Installed. No version info available.
tiktoken<1,>=0.7: Installed. No version info available.
typing-extensions>=4.7: Installed. No version info available.
zstandard: 0.23.0

[maxfriedrich]

I'm moving this into an issue because I'm pretty sure it's a bug