MIPod CLI is a command-line interface tool for deploying and serving Large Language Models (LLMs) with AMD GPU Pod scheduling.
- Schedule AMD GPU Pods for deploying LLMs.
- Manage and monitor your deployments.
- Easy-to-use CLI interface.
You can install MIPod CLI using pip:
pip install git+https://github.com/Hoipang/tokenvisor-cli
Once installed, you can use the mipod-cli command to interact with the tool. The main entry point for the CLI is defined in the mipod_cli.main module.
To get started, run:
mipod-cli --help
mipod-cli validate -f config.yaml
mipod-cli deploy -f config.yaml
This will display the available commands and options.