6d224861 came from a --no-checkout worktree whose empty index made the
commit drop every file except two. This restores 05b50d2b's full tree and
correctly adds stacks/stem95su/gdrive-sync.tf + the service-catalog stem95su
entry. Forward-only (parent=6d224861, no force-push); [ci skip] since the
live infra was never applied from the broken commit.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
CronJob stem95su-gdrive-sync (*/10) mounts the content PVC RW and
rclone-syncs the read-only Drive folder "claude" (stem claude/files) onto
it (rclone/rclone:1.74.3, scope=drive.readonly, empty-source guard +
--max-delete 25). ESO ExternalSecret stem95su-rclone <- Vault
secret/stem95su. Requires the GCP OAuth app published to Production or the
refresh token expires ~weekly.
Lands the gdrive-sync stack on master (it had landed on a feature branch
by accident on the shared devvm checkout).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
immich-ml at TTL=0 never unloaded models; a heavy OCR library job
inflated onnxruntime's CUDA arena to ~10.7GB and held it on the shared
time-sliced T4, starving llama-swap (qwen3-8b) so recruiter-responder
triage 502'd silently for hours (emails preserved unseen, no loss).
TTL=600 lets idle ad-hoc models (OCR, face) free VRAM while preloaded
CLIP/smart-search stays warm.
Docs: correct stale llama-cpp GPU notes (T4 is time-sliced, no VRAM
isolation; add qwen3-8b to model table), immich MODEL_TTL gotcha in
.claude/CLAUDE.md, and a post-mortem.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Phase 7 of the vision-LLM benchmark plan. Adds:
- docs/benchmarks/2026-05-10-vision-llm.md — curated report (TL;DR,
per-model analysis, top-N agreement, cost vs cloud APIs, sample
captions). Verdict: qwen3vl-4b for the request path (3.55 s p50,
100% parse, decisive top-N distro); qwen3vl-8b for caption polish.
- docs/benchmarks/benchmark-2026-05-10-1424.json — raw 300-row dump
for diff-checking against future runs.
- main.tf: -fa -> -fa on (b9085 llama.cpp removed the no-value form
of the flash-attention flag; without the value llama-server exits
before serving any request).
- llama-cpp.md architecture doc links the report so future operators
land on the deployed-and-evaluated model from one entry point.
300/300 calls, 0 parse errors, 33m32s wall on a single T4 with the
GPU exclusively allocated. immich-ml was scaled to 0 for the run
(node1 RAM constraint, not GPU - bumping node1 RAM is tracked as a
follow-up).
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Single Deployment of mostlygeek/llama-swap:cuda hot-swaps three
GGUF vision models (qwen3vl-8b, minicpm-v-4-5, qwen3vl-4b) at one
OpenAI-compat /v1 endpoint on Service llama-swap.llama-cpp.svc.
Idle TTL 10min so models unload between benchmark batches.
Storage: NFS-RWX from /srv/nfs-ssd/llamacpp (30Gi). One-shot
download Job pulls Q4_K_M GGUF + mmproj per model, creates stable
model.gguf / mmproj.gguf symlinks so the llama-swap config is
filename-agnostic, then warms the kernel page cache.
GPU: nvidia.com/gpu=1 = whole T4 — operator must scale immich-ml
to 0 during benchmark windows. wait_for_rollout=false so apply
doesn't block on GPU availability.
Initial use case: vision-LLM benchmark for instagram-poster
candidate scoring; future consumers (HA, agentic tooling) hit
the same endpoint via LiteLLM at the gateway.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
