llm/docs/aliases.md

(aliases)=

Model aliases

LLM supports model aliases, which allow you to refer to a model by a short name instead of its full ID.

Listing aliases

To list current aliases, run this:

llm aliases

Example output:

3.5         : gpt-3.5-turbo
chatgpt     : gpt-3.5-turbo
chatgpt-16k : gpt-3.5-turbo-16k
3.5-16k     : gpt-3.5-turbo-16k
4           : gpt-4
gpt4        : gpt-4
4-32k       : gpt-4-32k
ada         : ada-002 (embedding)

Add --json to get that list back as JSON:

llm aliases list --json

Example output:

{
    "3.5": "gpt-3.5-turbo",
    "chatgpt": "gpt-3.5-turbo",
    "chatgpt-16k": "gpt-3.5-turbo-16k",
    "3.5-16k": "gpt-3.5-turbo-16k",
    "4": "gpt-4",
    "gpt4": "gpt-4",
    "4-32k": "gpt-4-32k"
}

Adding a new alias

The llm aliases set <alias> <model-id> command can be used to add a new alias:

llm aliases set turbo gpt-3.5-turbo-16k

Now you can run the gpt-3.5-turbo-16k model using the turbo alias like this:

llm -m turbo 'An epic Greek-style saga about a cheesecake that builds a SQL database from scratch'

Removing an alias

The llm aliases remove <alias> command will remove the specified alias:

llm aliases remove turbo

Viewing the aliases file

Aliases are stored in an aliases.json file in the LLM configuration directory.

To see the path to that file, run this:

llm aliases path

To view the content of that file, run this:

cat "$(llm aliases path)"