| """Managed llama.cpp runtime. |
| |
| ``binaries`` resolves/downloads/verifies official llama.cpp release zips into |
| ``$HERMES_HOME/runtimes/llamacpp/<tag>/``; ``supervisor`` spawns and supervises one llama-server in |
| router mode (readiness is a touch generation, never health-200 alone); ``detect`` finds an |
| already-running llama-server (external or ours). |
| """ |
|
|
| from hermes_cli.local_runtime.binaries import ( |
| BinaryResolutionError, ensure_runtime_installed, resolve_assets, select_backend) |
| from hermes_cli.local_runtime.bootstrap import ensure_local_runtime, shutdown_local_runtime |
| from hermes_cli.local_runtime.context_policy import ( |
| FLOOR, growth_decision, initial_window, ladder, launch_args) |
| from hermes_cli.local_runtime.growth import ( |
| clear_window_override, load_window_overrides, maybe_grow_window, save_window_override) |
| from hermes_cli.local_runtime.detect import detect_server |
| from hermes_cli.local_runtime.endpoint import resolve_llamacpp_endpoint |
| from hermes_cli.local_runtime.estimator import ( |
| HardwareBudget, ctx_bytes, physics_check, profile_from_gguf) |
| from hermes_cli.local_runtime.gguf import read_gguf_header |
| from hermes_cli.local_runtime.hardware import probe_budget |
| from hermes_cli.local_runtime.presets import generate_presets |
| from hermes_cli.local_runtime.supervisor import LlamaServerSupervisor |
|
|
