> ## Documentation Index
> Fetch the complete documentation index at: https://www.edgee.ai/docs/llms.txt
> Use this file to discover all available pages before exploring further.
# LangChain
> Use Edgee with LangChain for building AI applications with chains, agents, and RAG.
Edgee's OpenAI-compatible API works with LangChain, allowing you to leverage LangChain's powerful
features like chains, agents, memory, and RAG while maintaining control over your LLM infrastructure.
## Installation
Using `uv` with inline dependencies (PEP 723):
```python theme={"dark"}
#!/usr/bin/env -S uv run
# /// script
# requires-python = ">=3.10"
# dependencies = [
# "langchain",
# "langchain-openai",
# ]
# ///
```
Or install manually:
```bash theme={"dark"}
pip install langchain langchain-openai
```
## Basic Usage
```python theme={"dark"}
from langchain_openai import ChatOpenAI
from langchain_core.messages import HumanMessage, SystemMessage
import os
# Initialize the LLM with Edgee endpoint
llm = ChatOpenAI(
base_url="https://edgee.io/v1",
api_key=os.getenv("API_KEY"),
model="mistral-small", # or any model available through Edgee
timeout=30.0,
)
# Simple chat
messages = [
SystemMessage(content="You are a helpful assistant."),
HumanMessage(content="What is LangChain?"),
]
response = llm.invoke(messages)
print(response.content)
```
## Command-Line Script
Complete script with argument parsing:
```python theme={"dark"}
#!/usr/bin/env -S uv run
# /// script
# requires-python = ">=3.10"
# dependencies = [
# "langchain",
# "langchain-openai",
# ]
# ///
from langchain_openai import ChatOpenAI
from langchain_core.messages import HumanMessage, SystemMessage
import os
import argparse
def main():
parser = argparse.ArgumentParser(description="LangChain with Edgee")
parser.add_argument("--model", type=str, default="mistral-small", help="Model name")
parser.add_argument("--message", type=str, required=True, help="Message to send")
parser.add_argument("--system", type=str, default="You are a helpful assistant.", help="System prompt")
args = parser.parse_args()
llm = ChatOpenAI(
base_url="https://edgee.io/v1",
api_key=os.getenv("API_KEY"),
model=args.model,
timeout=30.0,
)
messages = [
SystemMessage(content=args.system),
HumanMessage(content=args.message),
]
response = llm.invoke(messages)
print(response.content)
if __name__ == "__main__":
main()
```
### Usage Examples
```bash theme={"dark"}
# Basic usage (uses mistral-small by default)
uv run langchain_script.py --message "Tell me a joke"
# With custom model
uv run langchain_script.py --model "gpt-4" --message "Explain quantum computing"
# With custom system prompt
uv run langchain_script.py \
--model "mistral-small" \
--message "Write a haiku" \
--system "You are a creative poet"
```
## Compression & Tags via Headers
Control **Token Compression** and add tags for observability using the `default_headers` parameter:
```python theme={"dark"}
from langchain_openai import ChatOpenAI
import os
llm = ChatOpenAI(
base_url="https://edgee.io/v1",
api_key=os.getenv("API_KEY"),
model="gpt-5.2",
default_headers={
"x-edgee-compression-model": "claude",
"x-edgee-tags": "production,langchain,rag-pipeline",
}
)
```
**Available Headers:**
| Header | Type | Description |
| --------------------------- | ----------------------------------------------- | ------------------------------------------------------------------------ |
| `x-edgee-compression-model` | `"claude"`, `"opencode"`, `"cursor"`, `"codex"` | Compression bundle to apply (e.g. "claude" for Claude Token Compression) |
| `x-edgee-tags` | `string` | Comma-separated tags for analytics and filtering |
You can also enable compression per API key or Agent in the Edgee console. Headers override console settings.
## Advanced Features
### Chains
```python theme={"dark"}
from langchain_openai import ChatOpenAI
from langchain_core.prompts import PromptTemplate
from langchain_core.output_parsers import StrOutputParser
llm = ChatOpenAI(
base_url="https://edgee.io/v1",
api_key=os.getenv("API_KEY"),
model="mistral-small",
)
# Create a prompt template
prompt = PromptTemplate.from_template("Write a brief summary about {topic}")
# Create the chain using LCEL (LangChain Expression Language)
chain = prompt | llm | StrOutputParser()
# Run the chain
result = chain.invoke({"topic": "artificial intelligence"})
print(result)
```
### Streaming Responses
```python theme={"dark"}
llm = ChatOpenAI(
base_url="https://edgee.io/v1",
api_key=os.getenv("API_KEY"),
model="mistral-small",
streaming=True,
)
for chunk in llm.stream("Tell me a long story"):
print(chunk.content, end="", flush=True)
```
### Tags
You can add tags to your requests for analytics and filtering using the `default_headers` parameter:
```python theme={"dark"}
from langchain_openai import ChatOpenAI
import os
llm = ChatOpenAI(
base_url="https://edgee.io/v1",
api_key=os.getenv("API_KEY"),
model="mistral-small",
default_headers={
"x-edgee-tags": "production,user-123,langchain",
},
)
# All requests from this client will include these tags
response = llm.invoke("What is LangChain?")
```
Tags are comma-separated strings in the header. They help you categorize and filter requests in Edgee's analytics dashboard.
## Authentication
Edgee uses standard Bearer token authentication. Set your API key as an environment variable:
```bash theme={"dark"}
export API_KEY="sk-edgee-..."
```
The `api_key` parameter in `ChatOpenAI` automatically formats the header as:
```
Authorization: Bearer {api_key}
```
## Verify the connection
Send one request from the configured client, then open **Logs** in the same Edgee organization. Check the served model, provider, and key. For a CLI-launched coding agent, also inspect its session report.
If the request is missing, check that you launched through Edgee or saved the client’s gateway URL and Gateway API key. If it appears with an error, inspect the error before changing settings. See [Troubleshooting](/docs/troubleshooting) and [Gateway errors](/docs/llm-router/api-reference/errors).