wmantly de36ba8f6e feat(context): bump context window to 192K (196,608 tokens) with q4_0 KV cache

2026-09-22 01:50:58 +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

feat(context): bump context window to 192K (196,608 tokens) with q4_0 KV cache

2026-09-22 01:50:58 +00:00

start-server.sh

feat(context): bump context window to 192K (196,608 tokens) with q4_0 KV cache

2026-09-22 01:50:58 +00:00