tomoul models

List models — what's cached on disk, what's available in the catalog, and how much space each is taking.

Local (cached)

$ tomoul models
NAME                       SIZE     MODIFIED       QUANT
inkubalm-0.4b              240 MB   2 days ago     q8_k
baai/bge-m3                380 MB   3 hours ago    f16
openai/whisper-large-v3    1.5 GB   1 week ago     q8_0

This is the cache tomoul serve reads from. Defaults to ~/.cache/tomoul/; override via TOMOUL_CACHE_DIR or ~/.config/tomoul/config.toml.

Available

tomoul models --remote

Lists what's in the catalog. Cached entries are marked. Narrow with:

  • --filter llm
  • --filter embedding
  • --filter audio

Aliases

tomoul models --aliases

Prints the short-name → full-slug map.

Output formats

  • Default: human-readable table.
  • --json — JSON array (machine-readable, stable schema).
  • --format name — names only (good for pipelines).
# Nuke the whole cache:
tomoul models --format name | xargs -I{} tomoul rm {}
Last updated 13 May 2026Edit this page on GitHub