ommv0.2.148

omm run

Start a chat with a model already installed via omm — Ollama chats right here in the terminal, KoboldCpp and text-generation-webui start with the model loaded, GUI apps just open.

01 / 06

Overview

Reach for run once install has finished. With no name, run offers an interactive pick from what's installed; with a name, it resolves which runners that model is linked into and either drops you straight into a terminal chat (Ollama), launches the model already loaded (KoboldCpp, text-generation-webui), or opens the GUI app so you can pick it from its own local-models list. --engine overrides the automatic choice when a model is linked into more than one runner.

02 / 06

Options

Every flag this command accepts, and what it defaults to when you leave it out.

  • [name]—Default: picks interactively

    An installed model's filename (see omm list). Left out, run offers an interactive pick from what's installed.

  • --engine, -eNAMEDefault: auto-picks a linked engine

    Which linked runner to use: ollama, lmstudio, jan, koboldcpp, textgenwebui, anythingllm, or mstystudio. Left out, run picks one automatically.

  • --json—Default: off

    Accepted as a global flag, but run has no JSON output mode; omm prints a warning and ignores it.

  • --yes, -y—Default: off

    Accepted as a global flag, but run has no confirmation prompt to skip; omm prints a warning and ignores it.

  • --quiet, -q—Default: off

    Suppress background status or hint lines. Errors and the launch result still print.

  • --no-color—Default: off

    Disable colored terminal output.

  • --help—Default: off

    Print run's usage, arguments and options, then exit.

03 / 06

Examples

From a plain search to something you'd put in a script.

No name — pick interactively from what's installed.

$ omm run

Run a specific installed model, letting run pick the runner.

$ omm run qwen2.5-0.5b-instruct-q4_k_m.gguf

Run the same model through a specific runner instead of the automatic pick.

$ omm run qwen2.5-0.5b-instruct-q4_k_m.gguf --engine lmstudio

04 / 06

Recorded CLI run

Command recorded in this video

omm run --no-color

exit code: 1 · outcome: cancelled

Cancelled by the recorded keyboard input before any install, model run, benchmark, or upload.

Open the exact real-process transcript (.txt)

Extended documented terminal example

Format-accurate reproduction of the real startup banner (src/omm/cli.py:5538-5548) for a model genuinely installed on this dev machine — not a literal capture, since the chat itself is a live conversation this page can't script or reproduce.

06 / 06

If something goes wrong

Every message below is one this command actually prints. Find yours, read why it happened, then do the last line.

  1. totally-fake-model-xyz is not installed via omm. See `omm list`.
    why
    The name doesn't match anything omm has installed on this machine — it may be misspelled, or installed under a runner directly rather than through omm.
    what to do
    Run omm list to see the exact filename omm knows about, then pass that.
    source
    src/omm/cli.py:5533
  2. Ollama is not installed. Install it from https://ollama.com/download.
    why
    run picked (or was told to use) Ollama, but no Ollama installation was found on this machine.
    what to do
    Install Ollama from the link in the message, then retry.
    source
    src/omm/launcher.py:170
  3. `ollama run mistral-7b-instruct-v0.2.q4_k_m` exited with code 1. Try `omm link --engine ollama` to repair the model's Ollama link.
    why
    The model is linked into Ollama, but the ollama run itself failed or the link is stale.
    what to do
    Run omm link --engine ollama to re-verify and repair the link, then retry.
    source
    src/omm/launcher.py:177-183

Still stuck? Open an issue with the exact message you saw.