Use lms chat to talk to a local model directly in the terminal. This is handy for quick experiments or scripting.

[model] (optional) : string

Identifier of the model to use. If omitted, you will be prompted to pick one.

-p, --prompt (optional) : string

Send a one-off prompt and print the response to stdout before exiting

-s, --system-prompt (optional) : string

Custom system prompt for the chat

--stats (optional) : flag

Show detailed prediction statistics after each response

--ttl (optional) : number

Seconds to keep the model loaded after the chat ends (default: 3600)

lms chat

You will be prompted to pick a model if one is not provided.

lms chat my-model

Use -p to print the response to stdout and exit instead of staying interactive:

lms chat my-model -p "Summarize this release note"

lms chat my-model -s "You are a terse assistant. Reply in two sentences."

lms chat my-model --ttl 600

lms chat reads from stdin, so you can pipe content directly into a prompt:

cat my_file.txt | lms chat -p "Summarize this, please"