[Feature Request] Add customFetch and customHeaders support to ChatGoogleGenerativeAI

[Yuchaocheng]

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 very descriptive title to this issue.
• I searched the LangChain.js documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.
• I am sure that this is a bug in LangChain.js rather than my code.
• The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package).

Example Code

import { ChatGoogleGenerativeAI } from "@langchain/google-genai";

// ❌ Current: No way to customize fetch or headers
const model = new ChatGoogleGenerativeAI({
model: "gemini-2.0-flash",
apiKey: "my-api-key",
baseUrl: "https://my-proxy.example.com/api",
// These options are NOT supported currently:
// customFetch: myCustomFetch,
// customHeaders: { "Authorization": "Bearer my-token" },
});

// ✅ Expected: Should support customFetch and customHeaders
const modelWithCustomFetch = new ChatGoogleGenerativeAI({
model: "gemini-2.0-flash",
apiKey: "my-api-key",
baseUrl: "https://my-proxy.example.com/api",
customFetch: async (url, init) => {
// Custom logic here
return fetch(url, init);
},
customHeaders: new Headers({
"Authorization": "Bearer my-proxy-token",
}),
});

Error Message and Stack Trace (if applicable)

N/A - This is a feature request, not a bug.

Description

Problem:

The @google/generative-ai SDK (version 0.24.0+) supports customFetch and customHeaders options in RequestOptions, which allow users to:

1. Route requests through custom proxies with different authentication schemes
2. Add custom HTTP headers for enterprise/internal API gateways
3. Intercept and modify requests for debugging or logging

However, ChatGoogleGenerativeAI in @langchain/google-genai does not expose these options. Currently, only baseUrl and apiVersion are passed to the SDK:

// libs/providers/langchain-google-genai/src/chat_models.ts, constructor
{
  apiVersion: fields.apiVersion,
  baseUrl: fields.baseUrl,
}

Use Case:

Many enterprise users need to access Google Gemini models through internal proxy gateways that:

• Require Authorization: Bearer <token> header instead of x-goog-api-key
• Need custom headers for authentication, tracing, or rate limiting
• Require request/response logging via custom fetch

Current Workaround:

Users must patch globalThis.fetch, which is intrusive and affects all HTTP requests in the application.

Proposed Solution:

Add customFetch and customHeaders to GoogleGenerativeAIChatInput interface and pass them to the SDK:

// In GoogleGenerativeAIChatInput interface
customFetch?: typeof fetch;
customHeaders?: HeadersInit;

// In constructor, when creating client
{
  apiVersion: fields.apiVersion,
  baseUrl: fields.baseUrl,
  customFetch: fields.customFetch,
  customHeaders: fields.customHeaders,
}

This is a minimal, non-breaking change that unlocks significant flexibility for enterprise users.

System Info

Package: @langchain/google-genai
Version: 1.0.3
Underlying SDK: @google/generative-ai ^0.24.0

Platform: macOS / Linux / Windows (all affected)
Node version: 20+

Related source file:
https://github.com/langchain-ai/langchainjs/blob/main/libs/providers/langchain-google-genai/src/chat_models.ts