2026-04-30 - 2026-07-30
Overview
13 Issues created by 1 user
Opened
#1 Phase 1: GPU-Manager — Schritte 0–5 (ComfyUI systemd → mVoice load/unload → Broker MVP → wa.sh migrate → Queue → LRU-Eviction)
Opened
#2 Add a generic GPU-lease primitive (acquire/renew/release) for long-running async consumers (ComfyUI/FLUX)
Opened
#3 mGPUmanager: ollama not evictable + no game-awareness + OLLAMA_KEEP_ALIVE=24h starve the game/FLUX
Opened
#4 Broker reports per-consumer gpu_resident_mib=0 for externally-started consumers → eviction finds "no candidates" and never reclaims their VRAM
Opened
#5 Scheduler must evict ollama for higher-priority image leases (vram_managed guard defeats the manager's purpose)
Opened
#6 Broker must be the SOLE VRAM authority — close bypass gaps so out-of-band ollama holds get evicted (imagen starved by 24h keep-alive)
Opened
#7 STT big-model via GPU manager: mVoice eager-loads at startup + PWA bypasses the broker → big model OOM-crashes on a full GPU
Opened
#8 unload(): whisper-server systemd branch marks consumer unloaded without stopping the unit (latent VRAM-not-freed bug)
Opened
#9 STT forces German on every utterance: -l de hardcoded + language never forwarded, so English voice is translated to German
Opened
#10 STT glossary drift: consumers.yaml stt_prompt is hand-maintained and diverges from mVoice lexicon.yaml (missing mBrian → 'mBryan')
Opened
#11 comfyui's vram_budget_mib is 11000, measured usage is 4346 — the broker refuses image jobs that would fit comfortably
Opened
#12 comfyui's VRAM need is not a constant: it varies cold-vs-warm (8410 vs 6170 MiB) and with the requested resolution
Opened
#13 The broker can reclaim VRAM it did not allocate, but nothing stops a direct client taking it — a bare ollama run still OOM-kills whisper