wmantly ff29e72442 fix(memory): tune cache-ram to 12288MB to prevent host RAM OOM-kills on 48GB servers

2026-09-18 02:07:18 +00:00

..

build-nccl-llama.sh

Initial commit: Complete deployment scripts, power governor, systemd units, and architecture documentation for Turing multi-GPU LLM rig

2026-09-01 01:55:19 +00:00

ollama-proxy.py

fix(proxy): handle client aborts and broken pipe disconnections gracefully without 500 errors

2026-09-16 00:34:14 +00:00

start-server.sh

fix(memory): tune cache-ram to 12288MB to prevent host RAM OOM-kills on 48GB servers

2026-09-18 02:07:18 +00:00