Run the big model on the workstation, with admission control and the 1.7B as the floor #96

2026-08-02T20:30:27+02:00

claude commented

2026-08-02 20:30:27 +02:00

Closes Vikunja #489.

Acceptance criteria and quality gate are in TASK.md on this branch.
Review the review, not the diff — leave comments and the agent will apply them via task start 489.

Closes Vikunja #489. Acceptance criteria and quality gate are in `TASK.md` on this branch. Review the review, not the diff — leave comments and the agent will apply them via `task start 489`.

claude added 2 commits 2026-08-02 20:30:27 +02:00

mavgpud: build it with the rest, and ship the workstation config and unit (V-489) 8088ef9e00

make build now catches a broken supervisor on homesrv. deploy/mavgpud.json
carries the owner's gemma-4-12b line with the MTP draft model, passed to
llama-server untouched. The unit is a systemd user unit because sudo on the
workstation wants a password; lingering is the one command left to the owner.

docs: record that contention is KFD presence, not a VRAM threshold (V-489) cc423567e7

kami added 1 commit 2026-08-02 20:44:12 +02:00

mavgpud: turn gemma's thinking off in the chat template (V-489) eca41798bd

Owner's call, 02-08-2026. Without it the 12B spends the reply budget on
reasoning tokens and answers empty at low max_tokens. Verified on the box:
"Столица Франции?" now answers "Париж" with no reasoning_content.

kami merged commit 018f7a6f47 into master

2026-08-03 10:12:14 +02:00

Sign in to join this conversation.

2 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: kami/Maven#96