{
  "manifest_version": "1",
  "slug": "llm-router",
  "title": "LLM router",
  "headline": "Run a private LLM router for your team",
  "description": "Run LiteLLM on exe.dev: one OpenAI- and Anthropic-compatible endpoint for your team. Share the VM, and each person gets a key, a budget, and a log.",
  "category": "ai-apps",
  "tags": [
    "litellm",
    "llm gateway",
    "llm proxy",
    "openai compatible api",
    "ai gateway",
    "virtual keys",
    "llm budgets",
    "claude code",
    "codex",
    "docker compose",
    "postgresql"
  ],
  "url": "https://exe.dev/deploy/llm-router",
  "markdown_url": "https://exe.dev/deploy/llm-router.md",
  "upstream": {
    "website": "https://www.litellm.ai",
    "source": "https://github.com/BerriAI/litellm",
    "docs": "https://docs.litellm.ai/docs/simple_proxy",
    "license": "MIT (enterprise features under a separate license)"
  },
  "example_vm_name": "my-llm-router",
  "time": "15 minutes",
  "stack": [
    {
      "name": "LiteLLM 1.110",
      "detail": "The proxy and its admin UI, in Docker Compose",
      "data": "~/litellm/config.yaml"
    },
    {
      "name": "exe-auth",
      "detail": "A small program in front of LiteLLM that gives each person the VM is shared with a key",
      "port": 8000,
      "data": "~/litellm/exe-auth.py"
    },
    {
      "name": "PostgreSQL 18",
      "detail": "People, keys, budgets, and the request log",
      "data": "~/litellm/postgres"
    },
    {
      "name": "Redis 8",
      "detail": "Response cache, in memory, off until you turn it on"
    },
    {
      "name": "exe.dev LLM gateway",
      "detail": "Claude, GPT, Grok, Kimi, and GLM, without API keys"
    },
    {
      "name": "exe.dev login and VM tokens",
      "detail": "Say who each request is from; nothing else gets in"
    }
  ],
  "access": [
    {
      "url": "https://my-llm-router.exe.xyz/ui/",
      "who": "private",
      "what": "LiteLLM's admin UI, behind exe.dev login and your LiteLLM password"
    },
    {
      "url": "https://my-llm-router.exe.xyz/v1",
      "who": "private",
      "what": "The API, for anyone the VM is shared with, using an exe.dev VM token"
    }
  ],
  "delivers": [
    "LiteLLM at your VM's private URL, with you as its admin.",
    "Claude, GPT, Grok, Kimi, and GLM from exe.dev's LLM gateway behind one OpenAI- and Anthropic-compatible API.",
    "Share the VM, and each person gets their own key and $20 a month on their first request.",
    "Every request logged with its key, model, cost, and time; prompts aren't stored.",
    "Restarts after a reboot; nightly backups, with one checked."
  ],
  "deployable": true,
  "deploy": {
    "new_vm_url": "https://exe.dev/new?prompt=Set%20up%20LLM%20router%20on%20this%20VM%20by%20following%20the%20exe.dev%20recipe%20at%20https%3A%2F%2Fexe.dev%2Fdeploy%2Fllm-router.md%3Fvia%3Ddeploy%20from%20top%20to%20bottom.%0A%0AThe%20recipe%27s%20example%20VM%20name%20is%20my-llm-router.%20This%20VM%27s%20name%20is%20the%20output%20of%20%60hostname%60%3A%20use%20it%20wherever%20the%20recipe%20writes%20my-llm-router.%20Skip%20the%20step%20that%20creates%20the%20VM%3B%20you%20are%20on%20it.%0A%0AYou%20are%20done%20when%20all%20of%20these%20are%20true.%20Check%20each%20one%20yourself%3A%0A-%20LiteLLM%20at%20your%20VM%27s%20private%20URL%2C%20with%20you%20as%20its%20admin.%0A-%20Claude%2C%20GPT%2C%20Grok%2C%20Kimi%2C%20and%20GLM%20from%20exe.dev%27s%20LLM%20gateway%20behind%20one%20OpenAI-%20and%20Anthropic-compatible%20API.%0A-%20Share%20the%20VM%2C%20and%20each%20person%20gets%20their%20own%20key%20and%20%2420%20a%20month%20on%20their%20first%20request.%0A-%20Every%20request%20logged%20with%20its%20key%2C%20model%2C%20cost%2C%20and%20time%3B%20prompts%20aren%27t%20stored.%0A-%20Restarts%20after%20a%20reboot%3B%20nightly%20backups%2C%20with%20one%20checked.%0A%0ARead%20the%20whole%20recipe%2C%20including%20the%20closed%20details.%20For%20initial%20setup%2C%0Afollow%20%22Install%20by%20hand%22%20first%2C%20skipping%20VM%20creation%20and%20SSH%3A%20Docker%2C%20the%0Asecrets%2C%20the%20model%20list%2C%20the%20Compose%20stack%2C%20the%20model%20check%2C%20the%20owner%27s%0Aadmin%20account%2C%20the%20backup%20timer%2C%20and%20the%20backup%20check.%20Those%20checks%20are%0Aenough%3B%20don%27t%20install%20extra%20tools%20to%20test%20further.%20The%20walkthrough%20at%20the%0Atop%20is%20for%20the%20owner%20afterwards.%20Do%20not%20run%20Restore%20or%20Update.%20Keep%0Aexe-auth%20in%20front%20of%20LiteLLM%20on%20port%208000.%0AWork%20as%20exedev%20in%20~%2Flitellm.%20Never%20overwrite%20an%20existing%20.env%20or%0Aconfig.yaml%2C%20and%20never%20print%20the%20master%20key%20or%20a%20LiteLLM%20key.%0AMake%20the%20VM%27s%20owner%20the%20admin%2C%20using%20the%20email%20from%20the%20reflection%0Aintegration%3B%20if%20it%20prints%20nothing%2C%20ask%20the%20user%20for%20their%20exe.dev%20email.%0AGive%20the%20owner%20their%20sign-up%20link.%20Don%27t%20create%20VM%20tokens%2C%20integrations%2C%0Aor%20keys%20for%20other%20people%3A%20those%20are%20the%20owner%27s%20to%20hand%20out.%0AKeep%20test%20prompts%20to%20a%20word%20or%20two.%20Don%27t%20share%20the%20VM%20or%20make%20it%20public.%0AVerify%20docker.service%20is%20enabled%3B%20do%20not%20reboot.%0A%0AFinish%20with%20the%20URLs%20I%20should%20open.%0A",
    "terminal": "curl -fsSL https://exe.dev/deploy/llm-router/prompt.txt | ssh exe.dev new --prompt=/dev/stdin",
    "prompt": "Set up LLM router on this VM by following the exe.dev recipe at https://exe.dev/deploy/llm-router.md?via=deploy from top to bottom.\n\nThe recipe's example VM name is my-llm-router. This VM's name is the output of `hostname`: use it wherever the recipe writes my-llm-router. Skip the step that creates the VM; you are on it.\n\nYou are done when all of these are true. Check each one yourself:\n- LiteLLM at your VM's private URL, with you as its admin.\n- Claude, GPT, Grok, Kimi, and GLM from exe.dev's LLM gateway behind one OpenAI- and Anthropic-compatible API.\n- Share the VM, and each person gets their own key and $20 a month on their first request.\n- Every request logged with its key, model, cost, and time; prompts aren't stored.\n- Restarts after a reboot; nightly backups, with one checked.\n\nRead the whole recipe, including the closed details. For initial setup,\nfollow \"Install by hand\" first, skipping VM creation and SSH: Docker, the\nsecrets, the model list, the Compose stack, the model check, the owner's\nadmin account, the backup timer, and the backup check. Those checks are\nenough; don't install extra tools to test further. The walkthrough at the\ntop is for the owner afterwards. Do not run Restore or Update. Keep\nexe-auth in front of LiteLLM on port 8000.\nWork as exedev in ~/litellm. Never overwrite an existing .env or\nconfig.yaml, and never print the master key or a LiteLLM key.\nMake the VM's owner the admin, using the email from the reflection\nintegration; if it prints nothing, ask the user for their exe.dev email.\nGive the owner their sign-up link. Don't create VM tokens, integrations,\nor keys for other people: those are the owner's to hand out.\nKeep test prompts to a word or two. Don't share the VM or make it public.\nVerify docker.service is enabled; do not reboot.\n\nFinish with the URLs I should open.\n",
    "prompt_url": "https://exe.dev/deploy/llm-router/prompt.txt"
  },
  "updated": "2026-11-19",
  "verified": "2026-11-19",
  "related": [
    "open-webui",
    "librechat",
    "agent-secrets",
    "claude-code",
    "codex",
    "postgres"
  ],
  "alternatives": [
    "OpenRouter",
    "Portkey",
    "9router",
    "FreeLLMAPI"
  ],
  "files": [
    "https://exe.dev/deploy/llm-router/exe-auth.py"
  ]
}
