sglang 2

Spark Scope's mini window in the DeepSeek Harness sidebar: decode, prefill and every node at a glance

DSH plugin: inject per-model sampling parameters (top_p, top_k, min_p, repeat_penalty/repetition_penalty, presence_penalty, frequency_penalty) into every chat-completions request sent to an OpenAI-compatible gateway (llama.cpp, SGLang, vLLM).