gpu

GPU-aware execution layer for DeepSeek Harness: gpu_status / gpu_exec / gpu_run_bg tools, auto card selection, per-step GPU context

Safety-gated SSH development plugin for DeepSeek Harness and MCP, built for GPU and offline laboratory servers.

Manage local LLM inference servers in DSH: GPU registry, parameter profiles, VRAM validation, and tok/s benchmarks for llama.cpp, SGLang, and vLLM