Bridge by Wekoya / Guides / Use LangChain with Bridge

Use LangChain with Bridge

In short: Use LangChain's ChatOpenAI with base_url set to Bridge's API and your Bridge key. In Python, also pass use_responses_api=False so every request uses Chat Completions. Then any Bridge model ID works: GPT, Claude, Gemini or open-source, billed in naira.

Updated 8 November 2026

Python

pip install langchain-openai

import os
from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
    model="claude-sonnet-5-5",
    base_url="https://wekoya-api.wekoya.tech/v1",
    api_key=os.environ["WEKOYA_API_KEY"],
    use_responses_api=False,
)
print(llm.invoke("Hello").content)

Why use_responses_api=False: langchain-openai switches some models and options to OpenAI's Responses API on its own (for example GPT model IDs with tools). Bridge answers the Responses API too, but Chat Completions carries every Bridge option, so pinning it keeps behaviour the same on every model.

JavaScript

npm install @langchain/openai

import { ChatOpenAI } from "@langchain/openai"

const llm = new ChatOpenAI({
  model: "gemini-3.5-flash-lite",
  apiKey: process.env.WEKOYA_API_KEY,
  configuration: { baseURL: "https://wekoya-api.wekoya.tech/v1" },
})
console.log((await llm.invoke("Hello")).content)

Tool calling works with bind_tools (Python) or bindTools (JavaScript) on any model with tool calling. Checked against langchain-openai 1.7 and @langchain/openai 1.8 on 8 November 2026.

Create a free account