@adelie/gpu-model-serving

Stand up an OpenAI-compatible LLM endpoint on a rented or owned GPU box: pick the instance, fetch and quantize weights, serve with SGLang — speculative decoding, prefix caching and tool calling for agent workloads — then verify it and expose the port.