John Ogle 170a27310e feat(local-inference): add TTL support for automatic model unloading ...

Add globalTTL and per-model ttl options to llama-swap config,
allowing idle models to be automatically unloaded from memory.

2026-04-16 15:37:02 -07:00

.beads

…

.gitea/workflows

Remove beads/gastown/perles/dolt

2026-04-08 11:47:11 -07:00

home

fix(opencode): increase context/output limits for local model

2026-04-16 15:20:51 -07:00

machines

feat(local-inference): add TTL support for automatic model unloading

2026-04-16 15:37:02 -07:00

packages

feat(app-launcher): add /workout endpoint and URL arg support

2026-04-13 13:58:40 -07:00

roles

feat(local-inference): add TTL support for automatic model unloading

2026-04-16 15:37:02 -07:00

scripts

fix: add execute permission to rotate-wallpaper.sh

2026-03-05 07:33:15 -08:00

.gitattributes

Remove beads/gastown/perles/dolt

2026-04-08 11:47:11 -07:00

.gitignore

…

.goosehints

…

AGENTS.md

Remove beads/gastown/perles/dolt

2026-04-08 11:47:11 -07:00

flake.lock

Remove beads/gastown/perles/dolt

2026-04-08 11:47:11 -07:00

flake.nix

feat(local-inference): replace ollama with llama-swap + llama.cpp on zix790prors

2026-04-16 15:20:37 -07:00

renovate.json

Remove beads/gastown/perles/dolt

2026-04-08 11:47:11 -07:00

Description

No description provided

24 MiB

Languages

Nix 80.4%

Emacs Lisp 10.1%

Shell 6.2%

Python 3.3%