Calcis: LLM Cost Estimator szerző: Ryan Chen

Calcis shows you what an LLM API call will cost before you send it: right inside ChatGPT, Claude, and Gemini.

As you type a prompt, Calcis estimates the token count, predicts the output length, and shows the total cost in a small overlay near the input box. The estimate updates live as you type.

Features:
- Auto-detects which model you have selected
- Auto-detects effort/reasoning level
- Shows input tokens, predicted output tokens, and cost
- Shows cost at 1,000 requests/month for at-scale planning
- Works on chat.openai.com, chatgpt.com, claude.ai, gemini.google.com
- Non-intrusive overlay that can be collapsed or dismissed
- Your prompts are never stored or logged by the extension

Setup:
1. Get an API key at calcis.dev/dashboard
2. Click the Calcis extension icon and enter your API key
3. Open ChatGPT, Claude, or Gemini and start typing

Powered by the same estimation engine as calcis.dev.