I Built an Open-Source CLI to Compare LLM API Costs in Your Terminal (npx, Zero Install)

Source: Dev.to

If you’ve ever needed to compare costs between GPT‑4o, Claude Sonnet, Gemini, or any other LLM before committing to a model, you know the pain: juggling browser tabs, doing manual math, and relying on outdated blog posts.
llm-costs is a zero‑install CLI that does the heavy lifting instantly, counting tokens with the correct tokenizer and rendering a cost comparison table right in your terminal.

Why I built llm-costs

Every new LLM‑powered project used to start with the same ritual:

1. Open the Anthropic pricing page
2. Open the OpenAI pricing page
3. Open the Google AI pricing page
4. Try to compare apples to oranges (different tokenizers)
5. Do the math in my head or a spreadsheet
6. Realize the reference blog post is months out of date

There had to be a better way.

Quick demo

npx llm-costs "Build a REST API in Python" --compare

The CLI counts your prompt tokens using the actual tokenizer (tiktoken for OpenAI models, character‑based estimation for others) and prints a table such as:

Model                Input Cost   Output Cost   Total
──────────────────────────────────────────────────────
deepseek-chat        $0.00003     $0.00008      $0.00011
gemini-flash-2.0     $0.00005     $0.00020      $0.00025
claude-haiku-3-5    $0.00020     $0.00100      $0.00120
gpt-4o-mini          $0.00027     $0.00108      $0.00135
claude-sonnet-4-5    $0.00150     $0.00750      $0.00900
gpt-4o               $0.00375     $0.01500      $0.01875

Features

Zero install

Run the tool directly with npx or install it globally with npm—no manual setup required.

Multi‑provider support

• 17 models across 6 providers: Anthropic, OpenAI, Google, DeepSeek, Mistral, Cohere.

Auto‑updating prices

• Client‑side: On each run, the CLI checks ~/.llm-costs/pricing.json. If the file is older than 7 days, it fetches fresh data from GitHub (non‑blocking, 5 s timeout).
• Server‑side: A GitHub Actions workflow runs every Monday, pulls pricing from LiteLLM’s aggregate JSON, diffs the result, and opens a PR with a markdown table of changes.

Batch processing

Pipe a file of prompts to get total costs:

llm-costs batch prompts.txt

Budget guard

Set a cost ceiling for CI/CD pipelines:

llm-costs guard --max 0.10

Watch mode

Live‑refresh the cost table as you type your prompt.

MCP server mode

Integrate with Claude Desktop or any MCP‑compatible tool.

Price changelog

Track when costs changed:

llm-costs changelog --since 30d

Budget projections

Estimate future spend:

llm-costs budget --requests 10000

Installation & Usage

One‑shot, no install

npx llm-costs "your prompt here"

Global install

npm install -g llm-costs

Compare across all models

npx llm-costs "your prompt" --compare

Check a specific model

npx llm-costs "your prompt" --model claude-sonnet-4-5

Contributing

LLM pricing changes frequently, and the community can help keep llm-costs up to date. PRs to add new models, fix prices, or support additional providers are very welcome.

Links

• GitHub repository: ⭐
• npm package: