hypercli-cli 0.5.1

CLI for HyperCLI - GPU orchestration and LLM API

HyperCLI CLI

Command-line interface for HyperCLI - GPU orchestration and LLM API.

Installation

pip install hypercli-cli

This also installs the hypercli-sdk as a dependency.

Setup

Configure your API key:

hyper configure

Or set via environment:

export HYPERCLI_API_KEY=your_key

Or create ~/.hypercli/config:

HYPERCLI_API_KEY=your_key

Get your API key at hypercli.com/dashboard

Usage

Billing

hyper billing balance
hyper billing transactions
hyper billing balance -o json

LLM

# List models
hyper llm models

# Quick chat
hyper llm chat deepseek-v3.1 "Explain quantum computing"

# Interactive chat
hyper llm chat deepseek-v3.1

# With system prompt
hyper llm chat deepseek-v3.1 "Write a haiku" -s "You are a poet"

Jobs

# List jobs
hyper jobs list
hyper jobs list -s running

# Create a job
hyper jobs create nvidia/cuda:12.0 -g l40s -c "python train.py"

# Create and follow logs with TUI
hyper jobs create nvidia/cuda:12.0 -g h100 -n 8 -c "torchrun train.py" -f

# Get job details
hyper jobs get <job_id>

# Stream logs
hyper jobs logs <job_id> -f

# Watch GPU metrics
hyper jobs metrics <job_id> -w

# Cancel
hyper jobs cancel <job_id>

# Extend runtime
hyper jobs extend <job_id> 7200

User

hyper user

Output Formats

hyper jobs list -o json
hyper billing balance -o table

License

MIT