7.4 billion tokens per month. 34 free LLM providers. 635 free model endpoints. All behind one /v1 endpoint, plus any custom OpenAI-compatible endpoint. Smart routing, automatic failover, encrypted keys. Personal experimentation only.
TypeScript
import { defineConfig, loadEnv } from 'vite'
import react from '@vitejs/plugin-react'
import tailwindcss from '@tailwindcss/vite'
import path from 'path'
export default defineConfig(({ mode }) => {
const env = loadEnv(mode, path.resolve(__dirname, '..'), '')
const serverPort = env.PORT ?? process.env.PORT ?? 3001
return {
plugins: [react(), tailwindcss()],
base: process.env.VITE_BASE ?? '/',
envDir: path.resolve(__dirname, '..'),
define: {
__SERVER_PORT__: JSON.stringify(String(serverPort)),
},
resolve: {
alias: {
'@': path.resolve(__dirname, './src'),
},
},
server: {
proxy: {
// Force IPv4 — on Windows + Node 17+, `localhost` resolves to ::1 first,
// which can collide with wslrelay / Docker Desktop listeners on the same port.
'/api': `http://127.0.0.1:${serverPort}`,
'/v1': `http://127.0.0.1:${serverPort}`,
},
},
}
})