OpenAI

Connect hi-shell to OpenAI to use GPT models for shell command generation. GPT models provide excellent command generation quality with broad knowledge of shell commands across platforms.

Prerequisites

An OpenAI account
An API key from platform.openai.com/api-keys

Setup

1. Get an API Key

Sign up at platform.openai.com
Go to API Keys
Create a new secret key
Copy the key (starts with sk-)

2. Configure hi-shell

hi-shell --init
# Select "Cloud" → "OpenAI"
# Paste your API key
# Choose a model (e.g., "gpt-4o-mini")

Manual Configuration

Edit ~/.config/hi-shell/config.toml:

llm_provider = "Cloud"
cloud_provider = "OpenAI"
cloud_model = "gpt-4o-mini"
api_key = "sk-your-api-key-here"

Available Models

Model	Best For	Relative Cost
`gpt-4o-mini`	Fast, affordable, good quality	Low
`gpt-4o`	Best quality responses	Medium
`gpt-4.1-mini`	Latest fast model	Low
`gpt-4.1`	Latest flagship model	Medium

We recommend gpt-4o-mini as the default — it’s fast, affordable, and generates excellent shell commands.

Configuration Options

Option	Description
`cloud_model`	Model identifier
`api_key`	Your OpenAI API key

Pricing

OpenAI charges per token. For hi-shell usage:

Each command typically uses ~200-500 tokens
gpt-4o-mini: ~$0.0002 per command
gpt-4o: ~$0.001 per command

Very affordable for typical daily usage. Check openai.com/pricing for current rates.

Troubleshooting

“Invalid API key” Error

Verify your key at platform.openai.com/api-keys
Ensure you’re using an API key, not a ChatGPT Plus subscription
Check for extra whitespace

“Insufficient quota”

Add billing details in OpenAI platform settings
Check your usage limits

Organization Errors

If you’re part of multiple organizations, make sure you’re using the correct API key for the intended organization.