[FLEET-011] Local Model Pipeline and Fallback Chain #564

Closed
opened 2026-04-07 04:04:44 +00:00 by Rockachopa · 1 comment
Owner

Phase 4: Sovereignty | Requires: Phase 3 operational | Capacity cost: 50 | Innovation cost: 100

Every machine runs Ollama with 3+ models. Hard tasks route to bigger models. Easy tasks to small. All local. No cloud for core ops.

Acceptance

  • Ollama running on 3+ machines with 2+ models each
  • Fallback chain: if big model fails, try smaller
  • If all local fail, alert (no cloud fallback)
  • MILESTONE: A model ran locally. No cloud. No rate limits. No one can turn it off.
## Phase 4: Sovereignty | Requires: Phase 3 operational | Capacity cost: 50 | Innovation cost: 100 Every machine runs Ollama with 3+ models. Hard tasks route to bigger models. Easy tasks to small. All local. No cloud for core ops. ## Acceptance - [ ] Ollama running on 3+ machines with 2+ models each - [ ] Fallback chain: if big model fails, try smaller - [ ] If all local fail, alert (no cloud fallback) - [ ] MILESTONE: A model ran locally. No cloud. No rate limits. No one can turn it off.
Rockachopa added the phase-4projectfleet labels 2026-04-07 04:04:44 +00:00
gemini was assigned by allegro 2026-04-07 13:49:36 +00:00
Author
Owner

Completed. Script pushed to fleet/model-fallback.sh. PR #353 updated.

Completed. Script pushed to `fleet/model-fallback.sh`. PR #353 updated.
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Timmy_Foundation/timmy-home#564