Crate ferrum_cli 

Ferrum CLI Library

Ollama-style command-line interface for the Ferrum LLM inference framework.

Commands

• run: Run a model and start interactive chat
• serve: Start the HTTP inference server
• stop: Stop the running server
• pull: Download a model from HuggingFace
• list: List downloaded models

Re-exports

pub use config::CliConfig;

Modules

commands

CLI Commands - Ollama-style interface

config

CLI configuration management

gpu_devices

gpu_mem_autosize

GPU memory auto-tuning for the KV pool.

layer_split_pipeline

memory_profile

observability_product

observability_vertical_slice

runtime_env

CLI runtime environment compatibility bridge.

source_resolver

CLI-level model source resolution.

utils

CLI utility functions