A CLI tool that provides detailed information about AI language models, helping developers choose the right model for their needs.

Note: This project is now using the models.dev Database. Thanks for making it!

List models by popular providers

npx aidex --provider openai
npx aidex --provider anthropic
npx aidex --provider google

List all multimodal models that accept images as input:

npx aidex --input image --provider openai

Compare popular reasoning‑capable chat models:

npx aidex --compare "o3,opus4"

Find all models under $1 per million cache reads grouped by provider (note the additional --model filter which is required when using --group-by):

npx aidex --model gpt --group-by provider --sort-by cache_read_cost_per_token

The --sort-by flag accepts any numeric field returned by the API (after Aidex normalises the schema). The following fields are currently available:

• max_input_tokens – Maximum context window size.
• max_output_tokens – Maximum number of tokens the model can generate.
• input_cost_per_token – Cost per input token.
• output_cost_per_token – Cost per output token.
• cache_read_cost_per_token – Cost per cached-read token.
• cache_write_cost_per_token – Cost per cached-write token.

Example:

# Show the cheapest GPT-style models first
npx aidex --model gpt --sort-by input_cost_per_token

The --group-by option helps organise results into logical sections. It must be combined with either --model or --provider so that the search space is first narrowed before grouping.

Available grouping keys:

• provider – AI providers (OpenAI, Anthropic, etc.)
• type – Model capability buckets (Latest, Vision, etc.)
• mode – Model mode (Chat, Embedding, Rerank, …)
• series – Major model series (legacy vs latest, etc.)

Examples:

# Group every GPT-style model by provider and sort by cache-read cost
npx aidex --model gpt --group-by provider --sort-by cache_read_cost_per_token

# Show all OpenAI models grouped by type
npx aidex --provider openai --group-by type

# Combine convenience sort flags
npx aidex --mode chat --sort-cost --group-by mode

• X/Twitter: @kregenrek
• Bluesky: @kevinkern.dev

• Learn Cursor AI: Ultimate Cursor Course
• Learn to build software with AI: AI Builder Hub

• codefetch - Turn code into Markdown for LLMs with one simple terminal command
• instructa - Instructa Projects

unjs - for bringing us the best javascript tooling system