You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The Local · Qwen 4B tier is code-complete and CI-tested, but ONLY against a fake llama-server (packages/engine/test/local-llm-engine.e2e.test.ts uses fake-llama-server.mjs). No real end-to-end run is on record: real 2.5 GB model download, real llama-server b9601 startup, real translation/summary/coaching quality, offline operation, and the mid-session CLI→local fallback. Verify it for real on this Mac and file follow-ups for quality gaps.
Verification checklist (report N/N on this ticket)
Fresh download on this Mac: progress %, SHA-256 pass, .part resume after a forced interruption.
llama-server starts within budget; --reasoning-budget 0 honored (no multi-second think stalls).
Engine=local live session (network OFF after download): captions translated, summary + board generated, coaching rewrite works; subjective quality notes vs CLI on the same audio.
Fallback: start with engine=cli, kill the CLI process mid-session → router switches to local without dropping in-flight batches; status surfaced.
Checklist results on this ticket + follow-up tickets for anything below bar (esp. translation quality — if Qwen3-4B output is unusable, a model-upgrade/parameter follow-up).
Routing
agent/claude (PO direct — device-bound verification, manual §9/§12).
Part of #107. Depends on #108 (fast dev loop) for iteration comfort, not strictly blocking.
Goal
The Local · Qwen 4B tier is code-complete and CI-tested, but ONLY against a fake llama-server (
packages/engine/test/local-llm-engine.e2e.test.tsusesfake-llama-server.mjs). No real end-to-end run is on record: real 2.5 GB model download, realllama-server b9601startup, real translation/summary/coaching quality, offline operation, and the mid-session CLI→local fallback. Verify it for real on this Mac and file follow-ups for quality gaps.Where (reference, verified on main @ fa26d60)
packages/engine/src/pins.ts:24-31— pinnedQwen3-4B-Q4_K_M.gguf(2.497 GB, SHA-256) + llama.cppb9601src/host/local-tier.ts:90-173— binary + model acquisition,LazyLocalEnginepackages/engine/src/local-llm-engine.ts:101-167— server spawn args, health probe (60 s startup / 2 s per-poll / 30 s per-request)packages/engine/src/fallback-router.ts:35— CLI→local mid-session switchsrc/settings-sheet.ts:63-66Verification checklist (report N/N on this ticket)
.partresume after a forced interruption.llama-serverstarts within budget;--reasoning-budget 0honored (no multi-second think stalls).Deliverables
Checklist results on this ticket + follow-up tickets for anything below bar (esp. translation quality — if Qwen3-4B output is unusable, a model-upgrade/parameter follow-up).
Routing
agent/claude(PO direct — device-bound verification, manual §9/§12).Part of #107. Depends on #108 (fast dev loop) for iteration comfort, not strictly blocking.