langchain-dev-utils 0.1.11

A practical utility library for LangChain and LangGraph development

LangChain Dev Utils

This toolkit is designed to provide encapsulated utility functions for developers building applications with large language models using LangChain and LangGraph, helping developers work more efficiently.

Installation and Usage

1. Using pip

pip install -U langchain-dev-utils

2. Using poetry

poetry add langchain-dev-utils

3. Using uv

uv add langchain-dev-utils

Usage

1. Model Management ⭐

(1) ChatModel

from langchain_dev_utils import register_model_provider, load_chat_model
from langchain_qwq import ChatQwen
from dotenv import load_dotenv

load_dotenv()

# Register custom model providers
register_model_provider("dashscope", ChatQwen)
register_model_provider("openrouter", "openai", base_url="https://openrouter.ai/api/v1")

# Load models
model = load_chat_model(model="dashscope:qwen-flash")
print(model.invoke("Hello"))

model = load_chat_model(model="openrouter:moonshotai/kimi-k2-0905", temperature=0.7)
print(model.invoke("Hello"))

(2) Embedding

from langchain_dev_utils import register_embeddings_provider, load_embeddings
from langchain_siliconflow import SiliconFlowEmbeddings

register_embeddings_provider(
    "dashscope", "openai", base_url="https://dashscope.aliyuncs.com/compatible-mode/v1"
)

register_embeddings_provider("siliconflow", SiliconFlowEmbeddings)

embeddings = load_embeddings("dashscope:text-embedding-v4")
print(embeddings.embed_query("hello world"))

embeddings = load_embeddings("siliconflow:BAAI/bge-m3")
print(embeddings.embed_query("hello world"))

2. Message Utilities

from langchain_dev_utils import (
    convert_reasoning_content_for_ai_message,
    convert_reasoning_content_for_chunk_iterator,
    aconvert_reasoning_content_for_chunk_iterator,
    merge_ai_message_chunk,
    has_tool_calling,
    parse_tool_calling,
    message_format
)

# merge reasoning tags into content
msg = convert_reasoning_content_for_ai_message(ai_msg, think_tag=("<think>","</think>"))

# streaming (sync / async)
for chunk in convert_reasoning_content_for_chunk_iterator(model.stream("hi")):
    print(chunk.content, end="")

# re-assemble chunks
full = merge_ai_message_chunk(chunks)

# tool-call helpers
if has_tool_calling(msg):
    name, args = parse_tool_calling(msg, first_tool_call_only=True)

# pretty print mixed items
text = message_format(["text", "image", "note"], separator="\n", with_num=True)

3. Tool Enhancement

from langchain_dev_utils import human_in_the_loop, human_in_the_loop_async

@human_in_the_loop          # sync
@tool
def danger(x: int) -> str: ...

@human_in_the_loop_async    # async
@tool
async def danger_async(x: int) -> str: ...

4. Plan and Note

from langchain_dev_utils import (
    create_write_plan_tool,
    create_update_plan_tool,
    create_write_note_tool,
    create_query_note_tool,
    create_ls_tool,
    create_update_note_tool,
    PlanStateMixin,
    NoteStateMixin,
)

## define graph state
from langgraph.graph.message import MessagesState
class State(MessagesState, PlanStateMixin, NoteStateMixin):
    pass

tools=[
    create_write_plan_tool(),
    create_update_plan_tool(),
    create_write_note_tool(),
    create_query_note_tool(),
    create_ls_tool(),
    create_update_note_tool(),
]

5. Graph Pipeline

from langchain_dev_utils import sequential_pipeline, parallel_pipeline

sequential_pipeline(
    [
        subgraph1,
        subgraph2,
        subgraph3,
        subgraph4,
    ]
)

parallel_pipeline(
    [
        subgraph1,
        subgraph2,
        subgraph3,
        subgraph4,
    ]
)

Testing

All utility functions in this project have been tested. You can also clone the repository to run the tests:

git clone https://github.com/TBice123123/langchain-dev-utils.git

cd langchain-dev-utils

uv sync --group test

uv run pytest .

For more information, please refer to the following documents.