svayatta/loom

Distributed expert cache for large MoE inference: swarm-shared GGUF weights, node-local compute, gossip discovery, OpenAI-compatible API. One static Zig binary with Metal/Vulkan backends.

语言：Zig · GitHub 星标：6 · Fork：1

话题：distributed-systems · gguf · gossip-protocol · inference-engine · llm-inference · metal · mixture-of-experts · moe

在 GitHub 上查看原仓库