Gemini, without the Google SDK detour.
Request google/gemini-flash-1.5 from the OpenAI client you already ship. One OpenRouter key covers Pro for the heavy work and Flash for the fast, cheap lane.
$129/month SaaS. Bring your own model keys. No inference markup.
Three steps to connect.
Add an OpenRouter key
Gemini runs through OpenRouter, the same key that covers Anthropic and Meta models. Paste it in the ProxyLLM dashboard; it is encrypted with AES-256-GCM.
Point your SDK
Set the base URL to https://api.proxyllm.ai/v1 with your ProxyLLM key as the bearer. No Google AI SDK and no Vertex setup.
Request google/ models
google/gemini-pro-1.5 for the heavy work, google/gemini-flash-1.5 for the cheap fast lane. The slash in the model name tells the gateway to dispatch via OpenRouter.
Same client, different model name.
The google/ prefix sends the request through your OpenRouter key.
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api.proxyllm.ai/v1",
apiKey: "pk_live_…", // your ProxyLLM key
});
const r = await client.chat.completions.create({
model: "google/gemini-flash-1.5",
messages: [{ role: "user", content: "Summarize this thread." }],
});Run your AI workloads on your ChatGPT subscription.
ProxyLLM runs OpenAI's Codex for you, signed in with your own ChatGPT account. Your apps call one OpenAI-compatible endpoint and the work bills to your flat plan instead of per-token API pricing.
Skip the Vertex setup entirely.
Gemini runs on your own OpenRouter key, logged per request, no inference markup. For OpenAI-bound workloads, Codex Hosted bills them to your ChatGPT subscription instead of per-token pricing.