config: switch upgrader + report to deepseek, keep supervisor on glm #12

Merged
autonomic-bot merged 3 commits from config/deepseek-upgrader-model into main 2026-08-16 02:29:21 +00:00
Owner

upgrader.env sets LOOP_MODEL+REPORT_MODEL=tinfoil/deepseek-v4-pro so the weekly upgrader + report run on deepseek (no rolling balance limit). Supervisor stays on opencode-go/glm-5.2 (SUPERVISOR_MODEL default, not overridden). Subagents already bind deepseek via cc-ci repo config (verified this week: all 16 subagents deepseek). LOOP_TIER=zen kept so tier check passes; watchdog probe sends deepseek model name to zen endpoint -> 200 (not 429) -> resume immediately. dry-build passes. Root cause: 2026-08-14 run stalled on Insufficient balance (zen), sat unfinished 40h. Also documents session recovery in JOURNAL.md.

upgrader.env sets LOOP_MODEL+REPORT_MODEL=tinfoil/deepseek-v4-pro so the weekly upgrader + report run on deepseek (no rolling balance limit). Supervisor stays on opencode-go/glm-5.2 (SUPERVISOR_MODEL default, not overridden). Subagents already bind deepseek via cc-ci repo config (verified this week: all 16 subagents deepseek). LOOP_TIER=zen kept so tier check passes; watchdog probe sends deepseek model name to zen endpoint -> 200 (not 429) -> resume immediately. dry-build passes. Root cause: 2026-08-14 run stalled on Insufficient balance (zen), sat unfinished 40h. Also documents session recovery in JOURNAL.md.
autonomic-bot added 3 commits 2026-08-16 02:29:14 +00:00
The weekly /upgrade-all parent session and the /recipe-report session now
run on tinfoil/deepseek-v4-pro (LOOP_MODEL + REPORT_MODEL in upgrader.env).
The hourly supervisor stays on opencode-go/glm-5.2 (SUPERVISOR_MODEL default
in launch-supervisor.py, not overridden). Subagents already bind deepseek via
the cc-ci repo's opencode config (fix from 2026-08-10, verified this week:
all 16 subagents across both waves ran deepseek-v4-pro).

LOOP_TIER=zen is kept so the tier check passes; the watchdog's usage-limit
probe sends the deepseek model name to the zen endpoint, which returns 200
(not 429) → resume immediately — correct, since tinfoil has no rolling usage
limit to wait out. Verified the probe behaviour with a direct curl.

Root cause: the 2026-08-14 run stalled mid-recipe on 'Insufficient balance'
(opencode zen workspace balance exhausted), then sat unfinished for 40h while
the supervisor cron spun hourly unable to recover it. Deepseek (pay-per-use
API key) has no rolling balance limit, so this can't recur.

Also documents the session recovery in JOURNAL.md (the stalled run was
completed via a fresh scoped upgrader — the original 2.58M-token session was
unresumable: the inference endpoint silently drops the oversized request).
autonomic-bot merged commit a29b993160 into main 2026-08-16 02:29:21 +00:00
Sign in to join this conversation.
No Reviewers
No labels
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: recipe-maintainers/cc-ci-orchestrator#12