DEDUPED Hot AI repos, turned into use cases Get the Monday 5

How to call a running local model from your own scripts with Omarchy Local AI

Send OpenAI-compatible requests to a running model's gateway from your own code.

Credentials. The gateway key is the only protection on the gateway, so send it only to the gateway address and never paste it into shared code.

Who it is forA developer who already has scripts or tools that use the OpenAI API and wants them to run against a model on their own machine.

The problemExisting tools expect an OpenAI-style endpoint, and pointing them at a local model normally takes custom wiring for the address and key.

The outcomeYour scripts send requests to the model's gateway on this machine with the key, and get answers back from the local model.

What you need first

  • Omarchy with the Quattro shell and its bar
  • Docker, jq, curl, flock, sha256sum and sha1sum installed
  • A supported GPU, or an x86-64 CPU with AVX2 and enough RAM for the recipe
  • Enough free RAM and disk for the model you pick, as the panel lists them
  • Your Omarchy password for setup, since setup runs Omarchy's Sudoless Docker command, which is root-equivalent

Steps

  1. Start a model with the panel, or run bin/omarchy-local-ai run qwen38-27b-exl3-3bpw-rtx3090-sglang-tp1 nvidia:0 using a recipe your machine supports. (inferred, check the README)
  2. Open the model's page and copy its address under **this machine**; the address is hidden until you click it.
  3. Read the key from ~/.local/state/omarchy/local-ai/gateway.key.
  4. Send OpenAI-compatible requests to the gateway address with the key, and check that you get an answer.

Prompt for your agent

Using the repo 0xSero/omarchy-local-ai, start a validated model and send one OpenAI-compatible chat request from <script-path> to its gateway using the key in gateway.key. Expected output: a text reply from the local model. Verify that a request without the key is refused.

Caveat. The README says the key should go only to the gateway address on the model's Docker network, so calls from outside this machine will not reach it.

Share: X LinkedIn Hacker News Email Copy link

Disclaimer

Deduped lists open-source repositories that belong to other people. We do not own, maintain, audit, or vouch for any of them, and a repo appearing here is not an endorsement or a safety check. The use cases, steps, and prompts are written by AI and may be incomplete, outdated, or just wrong. Where a use case is marked "Seen in the wild", we link to public posts where someone described something similar, but we have not verified those posts and the use case itself is still our AI's own write-up. Use all of this at your own risk. Only run code, commands, or agents on systems and accounts you are authorized to use, and read the code and every command before you run it, especially anything that touches credentials, files, or the network. Nothing here is professional, security, legal, or financial advice. All trademarks and project names belong to their respective owners, and their use here does not imply affiliation. Content is provided as is, without warranties of any kind. Found a mistake, a safety problem, or something you want removed? Email [email protected] and we will look at it promptly. Full disclaimer page