API Reference

Integrate LLM cost estimation into your tools, pipelines, and IDE extensions. One endpoint, clean JSON, predictable pricing.

Get an API key from your dashboard →

Authentication

All API requests require a Bearer token in the Authorization header. Create API keys from your dashboard. API access requires Pro tier or above.

Authorization: Bearer calc_your_api_key_here

Endpoint

POSThttps://www.calcis.dev/api/v1/estimate

Request body

JSON object with the following fields:

Field	Type	Required	Description
`text`	string	Yes	The prompt text to estimate
`modelId`	string	Yes	Target model ID (e.g. "claude-sonnet-4-6", "gpt-4o")
`systemPrompt`	string	No	Optional system prompt (counted as input tokens)

Response

{
  "model": "claude-sonnet-4-6",
  "inputTokens": 847,
  "outputTokens": 523,
  "inputCost": 0.002541,
  "outputCost": 0.007845,
  "totalCost": 0.010386,
  "confidence": "high",
  "currency": "USD"
}

Field	Description
`model`	Model ID used for the estimate
`inputTokens`	Counted input tokens (prompt + system)
`outputTokens`	Predicted output tokens
`inputCost`	Input cost in USD
`outputCost`	Output cost in USD
`totalCost`	Total estimated cost in USD
`confidence`	"high" (regression model), "medium" (heuristic), or "low"
`currency`	Always "USD"

Examples

curl

curl -X POST https://www.calcis.dev/api/v1/estimate \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer calc_your_api_key_here" \
  -d '{
    "text": "Explain how transformer attention works",
    "modelId": "claude-sonnet-4-6"
  }'

JavaScript / Node.js

const response = await fetch(
  "https://www.calcis.dev/api/v1/estimate",
  {
    method: "POST",
    headers: {
      "Content-Type": "application/json",
      "Authorization": "Bearer calc_your_api_key_here",
    },
    body: JSON.stringify({
      text: "Explain how transformer attention works",
      modelId: "claude-sonnet-4-6",
    }),
  }
);

const estimate = await response.json();
console.log(estimate.totalCost); // 0.010386

Rate limits

60 requests per minute per API key. If you exceed this limit, the API returns 429 with a Retry-After header.

Error responses

Status	Meaning
`400`	Bad request (missing fields, unknown model)
`401`	Missing, invalid, or revoked API key
`429`	Rate limit exceeded
`500`	Internal error

Supported models

All models listed on the models page are supported. Pass the model ID as the modelId field.