Back to migration guides
From OpenRouter

Migrate from OpenRouter

Step-by-step guide to migrate from OpenRouter to LLM Gateway with minimal code changes

LLM Gateway provides a drop-in replacement for OpenRouter with OpenAI-compatible endpoints. Since both services follow the OpenAI API format, migration requires minimal changes to your existing code.

Quick Migration

Replace your OpenRouter configuration with LLM Gateway:

1- const baseURL = "https://openrouter.ai/api/v1";
2- const apiKey = process.env.OPENROUTER_API_KEY;
3+ const baseURL = "https://api.llmgateway.io/v1";
4+ const apiKey = process.env.LLM_GATEWAY_API_KEY;

Why Migrate to LLM Gateway?

Both OpenRouter and LLM Gateway offer robust LLM gateway solutions. Here's how they compare:

Feature OpenRouter LLM Gateway
OpenAI-compatible API Yes Yes
Multiple providers Yes (300+ models) Yes
Native AI SDK provider Yes Yes
Response caching Yes (prompt caching) Yes
Analytics dashboard Via third-party integrations Built-in
Cost tracking Yes Yes (per-request detail)
Provider key management Yes (BYOK) Yes (Pro)
Self-hosting option No Yes (AGPLv3)
Simpler API (no headers) Requires HTTP-Referer/X-Title Just Authorization
Anthropic-compatible API No Yes (/v1/messages)

For a detailed feature-by-feature comparison, see LLM Gateway vs OpenRouter.

Migration Steps

1. Get Your LLM Gateway API Key

Sign up at llmgateway.io/signup and create an API key from your dashboard.

2. Update Environment Variables

1# Remove OpenRouter credentials
2# OPENROUTER_API_KEY=sk-or-...
3

4# Add LLM Gateway credentials
5export LLM_GATEWAY_API_KEY=llmgtwy_your_key_here

3. Update Your Code

Using fetch/axios

1// Before (OpenRouter)
2const response = await fetch("https://openrouter.ai/api/v1/chat/completions", {
3  method: "POST",
4  headers: {
5    Authorization: `Bearer ${process.env.OPENROUTER_API_KEY}`,
6    "Content-Type": "application/json",
7    "HTTP-Referer": "https://your-site.com",
8    "X-Title": "Your App Name",
9  },
10  body: JSON.stringify({
11    model: "anthropic/claude-3-5-sonnet",
12    messages: [{ role: "user", content: "Hello!" }],
13  }),
14});
15

16// After (LLM Gateway)
17const response = await fetch("https://api.llmgateway.io/v1/chat/completions", {
18  method: "POST",
19  headers: {
20    Authorization: `Bearer ${process.env.LLM_GATEWAY_API_KEY}`,
21    "Content-Type": "application/json",
22  },
23  body: JSON.stringify({
24    model: "anthropic/claude-3-5-sonnet-20241022",
25    messages: [{ role: "user", content: "Hello!" }],
26  }),
27});

Using OpenAI SDK

1import OpenAI from "openai";
2

3// Before (OpenRouter)
4const client = new OpenAI({
5  baseURL: "https://openrouter.ai/api/v1",
6  apiKey: process.env.OPENROUTER_API_KEY,
7  defaultHeaders: {
8    "HTTP-Referer": "https://your-site.com",
9    "X-Title": "Your App Name",
10  },
11});
12

13// After (LLM Gateway)
14const client = new OpenAI({
15  baseURL: "https://api.llmgateway.io/v1",
16  apiKey: process.env.LLM_GATEWAY_API_KEY,
17});
18

19// Usage remains the same
20const completion = await client.chat.completions.create({
21  model: "anthropic/claude-3-5-sonnet-20241022",
22  messages: [{ role: "user", content: "Hello!" }],
23});

Using Vercel AI SDK

Both OpenRouter and LLM Gateway have native AI SDK providers, making migration straightforward:

1import { generateText } from "ai";
2

3// Before (OpenRouter AI SDK Provider)
4import { createOpenRouter } from "@openrouter/ai-sdk-provider";
5

6const openrouter = createOpenRouter({
7  apiKey: process.env.OPENROUTER_API_KEY,
8});
9

10const { text } = await generateText({
11  model: openrouter("gpt-5.2"),
12  prompt: "Hello!",
13});
14

15// After (LLM Gateway AI SDK Provider)
16import { createLLMGateway } from "@llmgateway/ai-sdk-provider";
17

18const llmgateway = createLLMGateway({
19  apiKey: process.env.LLMGATEWAY_API_KEY,
20});
21

22const { text } = await generateText({
23  model: llmgateway("gpt-5.2"),
24  prompt: "Hello!",
25});

Model Name Mapping

Most model names are compatible, but here are some common mappings:

OpenRouter Model LLM Gateway Model
gpt-5.2 gpt-5.2 or openai/gpt-5.2
claude-opus-4-5-20251101 claude-opus-4-5-20251101 or anthropic/claude-opus-4-5-20251101
gemini/gemini-3-flash-preview gemini-3-flash-preview or google-ai-studio/gemini-3-flash-preview
bedrock/claude-opus-4-5-20251101 claude-opus-4-5-20251101 or aws-bedrock/claude-opus-4-5-20251101

Check the models page for the full list of available models.

Streaming Support

LLM Gateway supports streaming responses identically to OpenRouter:

1const stream = await client.chat.completions.create({
2  model: "anthropic/claude-3-5-sonnet-20241022",
3  messages: [{ role: "user", content: "Write a story" }],
4  stream: true,
5});
6

7for await (const chunk of stream) {
8  process.stdout.write(chunk.choices[0]?.delta?.content || "");
9}

Additional Benefits

After migrating to LLM Gateway, you get access to:

  • Response Caching: Automatic caching for identical requests to reduce costs
  • Detailed Analytics: Per-request cost tracking, latency metrics, and usage patterns
  • Provider Key Management: Use your own API keys for providers (Pro plan)
  • Self-Hosting: Deploy LLM Gateway on your own infrastructure

Full Comparison

Want to see a detailed breakdown of all features? Check out our LLM Gateway vs OpenRouter comparison page.

Need Help?