# Building GoRoute: An Ultra-Lightweight (<15MB RAM) LLM Gateway in Go

> Source: <https://dev.to/zius/building-goroute-an-ultra-lightweight-15mb-ram-llm-gateway-in-go-5dnj>
> Published: 2026-10-18 03:32:18+00:00

Most developers using terminal coding agents like **Claude Code**, **OpenCode**, **Codex**, or **Cursor** end up running an LLM router in the background to handle rate limits, multiple accounts, and model failovers.

Popular tools like OmniRoute and LiteLLM work, but running a full Node.js or Python runtime 24/7 just to proxy HTTP requests consumes 150MB to 300MB of RAM and introduces cold-start latency.

I wanted something that idles at `< 6MB RAM`, starts instantly, and never panics. So I built **GoRoute**.

[GoRoute](https://github.com/ziuus/goroute) is a zero-dependency, compiled LLM gateway and multi-agent router written in Go.

It acts as a local reverse proxy (`http://127.0.0.1:8128`) sitting between your developer agents and AI providers.

```
+-------------+       +-------------------+       +-----------------------+
| Claude Code | ----> |                   | ----> | Google Antigravity    |
| OpenCode    | ----> | GoRoute (Go Proxy)| ----> | OpenAI / Anthropic    |
| Codex/Cursor| ----> |   (<15MB RAM)     | ----> | Custom Endpoints      |
+-------------+       +-------------------+       +-----------------------+
```

`/v1/chat/completions`) or Anthropic Messages API (`/v1/messages`) seamlessly.

```
   goroute setup all
goroute import-omniroute
```

You can install GoRoute directly via curl:

```
curl -fsSL https://raw.githubusercontent.com/ziuus/goroute/main/install.sh | bash
```

Or run via npm:

```
npx goroute setup all
```

Once installed, check status:

```
goroute status
```

Check out the GitHub repo here: [https://github.com/ziuus/goroute](https://github.com/ziuus/goroute)

I'd love to hear your feedback, feature requests, or questions!