Model Manager Runtime

Backend-selected model manager for the HTTP server.

This thin shim keeps the HTTP server code importing one stable manager type while build-time backend selection decides whether that implementation comes from the Vulkan runtime or the Apple Silicon Metal runtime.

2 exports 0 methods src/server/model_manager_runtime.zig

2 exports shown

constant

LoadSpec

pub const LoadSpec = impl.LoadSpec

Specification describing which model to load: a filesystem path, an optional managed-catalog ID, and an optional context-length override.

src/server/model_manager_runtime.zig:15

constant

ModelManager

pub const ModelManager = impl.ModelManager

Thread-safe manager for the currently active model and inference engine; handles loading, hot-swapping, catalog queries, and VRAM budget enforcement.

src/server/model_manager_runtime.zig:17