When langgraph's stream_mode includes custom, the cached results do not contain custom data.

[AstonyCat]

Checked other resources

• This is a bug, not a usage question. For questions, please use the LangChain Forum (https://forum.langchain.com/).
• I added a clear and detailed title that summarizes the issue.
• I read what a minimal reproducible example is (https://stackoverflow.com/help/minimal-reproducible-example).
• I included a self-contained, minimal example that demonstrates the issue INCLUDING all the relevant imports. The code run AS IS to reproduce the issue.

Example Code

from langgraph.config import get_stream_writer
from typing import TypedDict

from langgraph.cache.memory import InMemoryCache
from langgraph.checkpoint.memory import InMemorySaver
from langgraph.graph import StateGraph
from langgraph.types import CachePolicy


class State(TypedDict):
    x: int
    result: int


def expensive_node(state: State) -> State:
    print('NOT CACHED')
    writer = get_stream_writer()
    writer({
        "token_usage": "{'prompt_tokens': 10, 'completion_tokens': 20, 'total_tokens': 30}",
    })
    return {"result": state["x"] * 2}


def cache_key(state: State):
    return str(state["x"])


graph = (
    StateGraph(State)
    .add_node(expensive_node, cache_policy=CachePolicy(key_func=cache_key))
    .set_entry_point("expensive_node")
    .set_finish_point("expensive_node")
    .compile(cache=InMemoryCache(), checkpointer=InMemorySaver())
)

CONFIG = {'configurable': {'thread_id': 1}}
print(graph.invoke({"x": 5}, config=CONFIG, stream_mode=["updates","custom"]))
print(graph.invoke({"x": 5}, config=CONFIG, stream_mode=["updates","custom"]))

Error Message and Stack Trace (if applicable)

NOT CACHED
[('custom', {'token_usage': "{'prompt_tokens': 10, 'completion_tokens': 20, 'total_tokens': 30}"}), ('updates', {'expensive_node': {'result': 10}})]
[('updates', {'expensive_node': {'result': 10}, '__metadata__': {'cached': True}})]

Description

When langgraph's stream_mode includes custom, the cached results do not contain custom data.

System Info

System Information

OS: Windows
OS Version: 10.0.26100
Python Version: 3.11.13 | packaged by Anaconda, Inc. | (main, Jun 5 2025, 13:03:15) [MSC v.1929 64 bit (AMD64)]

Package Information

langchain_core: 0.3.76
langchain: 0.3.27
langchain_community: 0.3.29
langsmith: 0.4.27
langchain_chroma: 0.2.4
langchain_graph_retriever: 0.8.0
langchain_milvus: 0.2.1
langchain_openai: 0.3.33
langchain_postgres: 0.0.15
langchain_tavily: 0.2.11
langchain_text_splitters: 0.3.11
langgraph_sdk: 0.2.6

Optional packages not installed

langserve

Other Dependencies

aiohttp: 3.12.15
aiohttp<4.0.0,>=3.8.3: Installed. No version info available.
astrapy>=2.0.0;: Installed. No version info available.
async-timeout<5.0.0,>=4.0.0;: Installed. No version info available.
asyncpg>=0.30.0: Installed. No version info available.
backoff>=2.2.1: Installed. No version info available.
beautifulsoup4>=4.12.3;: Installed. No version info available.
cassio>=0.1.10;: Installed. No version info available.
chromadb>=0.5.23;: Installed. No version info available.
chromadb>=1.0.9: Installed. No version info available.
dataclasses-json<0.7,>=0.6.7: Installed. No version info available.
gliner==0.2.13;: Installed. No version info available.
graph-retriever: 0.8.0
httpx-sse<1.0.0,>=0.4.0: Installed. No version info available.
httpx<1,>=0.23.0: Installed. No version info available.
httpx>=0.25.2: Installed. No version info available.
httpx>=0.28.1;: Installed. No version info available.
immutabledict>=4.2.1: Installed. No version info available.
jsonpatch<2.0,>=1.33: Installed. No version info available.
keybert>=0.8.5;: Installed. No version info available.
langchain-anthropic;: Installed. No version info available.
langchain-astradb>=0.6.0;: Installed. No version info available.
langchain-aws;: Installed. No version info available.
langchain-azure-ai;: Installed. No version info available.
langchain-chroma>=0.2.0;: Installed. No version info available.
langchain-cohere;: Installed. No version info available.
langchain-community;: Installed. No version info available.
langchain-community>=0.3.14;: Installed. No version info available.
langchain-core<0.4.0,>=0.2.13: Installed. No version info available.
langchain-core<1.0.0,>=0.3.72: Installed. No version info available.
langchain-core<1.0.0,>=0.3.76: Installed. No version info available.
langchain-core<2.0.0,>=0.3.75: Installed. No version info available.
langchain-core>=0.3.29: Installed. No version info available.
langchain-core>=0.3.60: 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-perplexity;: Installed. No version info available.
langchain-text-splitters<1.0.0,>=0.3.9: Installed. No version info available.
langchain-together;: Installed. No version info available.
langchain-xai;: Installed. No version info available.
langchain<2.0.0,>=0.3.27: Installed. No version info available.
langsmith-pyo3>=0.1.0rc2;: Installed. No version info available.
langsmith>=0.1.125: Installed. No version info available.
langsmith>=0.1.17: Installed. No version info available.
langsmith>=0.3.45: Installed. No version info available.
networkx>=3.4.2: Installed. No version info available.
numpy<3,>=1.21: Installed. No version info available.
numpy>=1.26.0;: Installed. No version info available.
numpy>=1.26.2;: Installed. No version info available.
numpy>=2.1.0;: Installed. No version info available.
openai-agents>=0.0.3;: Installed. No version info available.
openai<2.0.0,>=1.104.2: Installed. No version info available.
opensearch-py>=2.8.0;: Installed. No version info available.
opentelemetry-api>=1.30.0;: Installed. No version info available.
opentelemetry-exporter-otlp-proto-http>=1.30.0;: Installed. No version info available.
opentelemetry-sdk>=1.30.0;: Installed. No version info available.
orjson>=3.10.1: Installed. No version info available.
orjson>=3.9.14;: Installed. No version info available.
packaging>=23.2: Installed. No version info available.
pgvector<0.4,>=0.2.5: Installed. No version info available.
psycopg-pool<4,>=3.2.1: Installed. No version info available.
psycopg<4,>=3: Installed. No version info available.
pydantic-settings<3.0.0,>=2.10.1: Installed. No version info available.
pydantic<3,>=1: Installed. No version info available.
pydantic<3.0.0,>=2.7.4: Installed. No version info available.
pydantic>=2.10.4: Installed. No version info available.
pydantic>=2.7.4: Installed. No version info available.
pymilvus: 2.6.1
pytest>=7.0.0;: Installed. No version info available.
PyYAML>=5.3: Installed. No version info available.
requests: 2.32.5
requests-toolbelt>=1.0.0: Installed. No version info available.
requests<3,>=2: Installed. No version info available.
requests<3,>=2.32.5: Installed. No version info available.
requests>=2.0.0: Installed. No version info available.
rich>=13.9.4;: Installed. No version info available.
spacy>=3.8.4;: Installed. No version info available.
SQLAlchemy<3,>=1.4: Installed. No version info available.
sqlalchemy<3,>=2: Installed. No version info available.
tenacity!=8.4.0,<10,>=8.1.0: 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>=4.12.2: Installed. No version info available.
typing-extensions>=4.7: Installed. No version info available.
vcrpy>=7.0.0;: Installed. No version info available.
zstandard>=0.23.0: Installed. No version info available.

[ArjunJagdale]

should cached runs re-emit custom data (e.g. token usage) as-is, or is it expected that such data only appears during live execution?
i can confirm the repro is valid — cached invocations return only updates, dropping custom events written via get_stream_writer()

[qingshanyuluo]

Hello LangChain Team,

I've carefully reviewed this issue and have a clear understanding of the root cause. I would like to volunteer to work on a fix.

My analysis is that the current caching mechanism is designed to store and retrieve only the final state update from a node's execution. It doesn't account for the sequence of stream events (like custom data) that are generated during that execution. When a cache hit occurs, the node's function is skipped, and so is the logic that emits the custom events.

My proposed approach to solve this would be:

Enhance the Cached Value: Instead of just storing the final state dictionary, the cache entry should store a more comprehensive object. This object would contain both the state_update and a list of all stream_events captured during the initial run.

On Cache Miss (Recording Phase): During the first execution of the node, we will not only yield the stream events to the user but also collect them into a list. This list, along with the final state update, will be saved to the cache.

On Cache Hit (Replay Phase): When a cache hit occurs, the logic will retrieve the full cached object. It will then iterate through the saved stream_events list and yield each event, effectively "replaying" the original stream. The metadata: {'cached': True} flag will still be injected into the updates event to maintain transparency.

This approach ensures that the user experience with streaming is consistent, regardless of whether the data is coming from a live execution or the cache.

If this plan sounds reasonable, could you please assign this issue #6265 to me? I'm ready to start working on a pull request.

Thank you!