Repository navigation
dense GGUF CUDA-init fallback to CPU is silent in text mode #4977
Description
PROPOSE-TICKET from la-0.71.1a (inbox 13:39:22Z), relayed by the cop. Owner: la-0.71.1a.
Done when: a dense GGUF run on a cuda build whose CUDA init fails prints one stderr line naming the CPU fallback, without --verbose, as the Qwen3.5 path already does.
Falsifier: CUDA_VISIBLE_DEVICES= apr run <dense gguf> --prompt x -n 8 prints no fallback line today (JSON says fell_back true; text mode is silent).
Placement: milestone 0.71.1; E2/E3 are full. Held unlinked and flagged, queued behind #4971 for the first E2 or E3 slot, by MOVE.
Agent: aprender-triage-worker
Activity
noahgift commentedon Oct 14, 2026
merged into #5312 (master ticket for epic #3994; triage rule 7b, 2026-10-14). It is a checklist row there; reopen this issue when work on it starts or its PR opens.
noahgift commentedon Oct 14, 2026
Moved: this sub-ticket is now a row in #5325 (apr on GPU: refuse by name before OOM, never fall back to CPU silently), split out of #5312 by theme.
test(serve): source-scan that the CUDA-init fallback line is not behi…