APIMaster.ai
APIMaster.ai Docs
中文Quick startAPIMaster.aiMarketplace

LangChain + APIMaster.ai

Use APIMaster.ai OpenAI-compatible API in LangChain instead of the official OpenAI quickstart key.

LangChain is a popular framework for LLM applications. APIMaster.ai exposes an OpenAI-compatible API — set model_provider="openai" and point base_url at APIMaster.

Get an API Key first. Copy the exact model id from the marketplace (e.g. gpt-5.4, claude-sonnet-4-6).

Prerequisites

  1. Python 3.10+ (3.11+ recommended).
  2. An APIMaster API Key from the console.
  3. A target model id from the marketplace.

Step 1 — Install dependencies

pip install langchain langchain-openai httpx

Step 2 — Create the sample file

Create apimaster_quickstart.py:

import httpx

from langchain.agents import create_agent
from langchain.chat_models import init_chat_model


APIMASTER_API_KEY = "your APIMaster.ai key"
APIMASTER_BASE_URL = "https://apimaster.ai/v1"
MODEL_NAME = "gpt-5.4"


def get_weather(city: str) -> str:
    """Get weather for a given city."""
    return f"It's always sunny in {city}!"


def main() -> None:
    model = init_chat_model(
        MODEL_NAME,
        model_provider="openai",
        api_key=APIMASTER_API_KEY,
        base_url=APIMASTER_BASE_URL,
        http_client=httpx.Client(trust_env=False, timeout=60),
        timeout=60,
    )

    agent = create_agent(
        model=model,
        tools=[get_weather],
        system_prompt="You are a helpful assistant",
    )

    result = agent.invoke(
        {"messages": [{"role": "user", "content": "What's the weather in San Francisco?"}]}
    )

    print(result["messages"][-1].content_blocks)


if __name__ == "__main__":
    main()

You can also download the sample script and paste your key before running.

Step 3 — Run

python apimaster_quickstart.py

On success you should see output similar to:

[{'type': 'text', 'text': "It's always sunny in San Francisco!"}]

The agent calls the get_weather tool and returns the final reply.

Key settings

APIMaster OpenAI-compatible base URL:

https://apimaster.ai/v1

Core LangChain configuration:

model = init_chat_model(
    "gpt-5.4",
    model_provider="openai",
    api_key=APIMASTER_API_KEY,
    base_url="https://apimaster.ai/v1",
)
Parameter Value
model_provider "openai" (OpenAI-compatible protocol)
base_url https://apimaster.ai/v1
Model name Marketplace model id

GPT example: MODEL_NAME = "gpt-5.4"
Claude example: MODEL_NAME = "claude-sonnet-4-6"

Proxy issues

If HTTP_PROXY / HTTPS_PROXY is set locally, you may hit SSL or connection errors. Pass:

http_client=httpx.Client(trust_env=False, timeout=60)

This stops httpx from picking up system proxy env vars — useful for quick local tests. Configure proxies explicitly in production if needed.

Safer key handling

For real projects, use environment variables instead of hard-coding keys:

import os

APIMASTER_API_KEY = os.environ["APIMASTER_API_KEY"]
APIMASTER_BASE_URL = "https://apimaster.ai/v1"

export APIMASTER_API_KEY="your key"
python apimaster_quickstart.py

Windows PowerShell:

$env:APIMASTER_API_KEY="your key"
python apimaster_quickstart.py

Troubleshooting

Symptom Fix
401 / Invalid API Key Verify key is complete and enabled in console
404 / model not found MODEL_NAME must match marketplace model id exactly
SSL / timeout Try trust_env=False; check firewall/proxy
ModuleNotFoundError Run pip install langchain langchain-openai httpx

Checklist

  • Installed langchain, langchain-openai, httpx
  • base_url = https://apimaster.ai/v1
  • MODEL_NAME from marketplace
  • API Key set (code or env var)
  • apimaster_quickstart.py runs and prints agent output

See also