diff options
| author | Henry J Webster <hwebs@hwebs.info> | 2026-08-04 00:02:54 -0500 |
|---|---|---|
| committer | Henry J. Webster <hwebs@hwebs.info> | 2026-08-04 16:05:20 -0500 |
| commit | df960b2ed92f9e45fc4984d19ab65c55074bb1d7 (patch) | |
| tree | 702ad999126a55f93122277c986a1bd6974a47be /common/AGENTS.md | |
| parent | 8f9fbf2f4a269eae143c9e279afc6608ce64b7d0 (diff) | |
kusanagi/ollama: double default context, add 1h keep-alive
The 12B model at 32k context uses only 7.7 GB of the 7900 XT's 20 GB, so
there is ample VRAM headroom for 64k with the q8_0 KV cache — agentic
coding tools are context-hungry and 32k is the practical bottleneck.
The keep-alive stops the default 5-minute idle unload from adding a
reload stall to every resumed session on a single-user workstation.
Update flake to get newer packages.
Assisted-by: Claude Code:claude-fable-5
Diffstat (limited to 'common/AGENTS.md')
0 files changed, 0 insertions, 0 deletions