[calcis]
EstimatorModelsPricing
Docs
Cost Calculator
Numbers in, total out
Session Simulator
Multi-turn cost compounding
Conversation Analyser
Upload a context file
Workload Calculators
Chatbot, RAG, embedding...
Token Counters
Per-provider tokenisers
API Reference
POST /api/v1/estimate
How it works

Product

  • Estimator
  • Calculator
  • Workload calculators
  • Token tools
  • Converters
  • Cost badges
  • Session

Resources

  • All models
  • Compare
  • Pricing
  • Q2 2026 report
  • API docs
  • Changelog
  • Release notes
  • RSS feed

Company

  • How it works
  • Methodology
  • Why Calcis
  • GitHub Action
  • Support
  • calcis.dev@gmail.com

Legal

  • Privacy
  • Terms
  • Accessibility
  • Sub-processors
Calcis - Know what your LLM calls cost before you make them. | Product HuntCalcis approved on SaaSHub
© 2026 Calcis · LLM cost estimates before you run anything.Calcis produces estimates, not invoices. See the accuracy disclaimer.

Estimate before you spend

Paste a prompt, pick a model, see the cost. Token-exact inputs, predicted outputs with a confidence band.

You have hit the 50-estimate monthly soft cap.

Anonymous use stays free. Sign up free for a 200-estimate monthly allowance and the ability to save your estimates. Upgrade to Pro for the LLM-assisted predictor (the most accurate layer in the stack), the Bayesian P10 to P90 confidence band, and an unlimited per-tier quota.

  • Sign up free, no card needed
  • Pro adds the LLM-assisted predictor (Precise mode)
  • Cancel anytime, full refund within 30 days
Sign up freeGo Pro

Prompt

Prompts aren't stored. OpenAI counts run locally; Claude and Gemini are sent to provider count-only endpoints. Privacy details

Target model

INPUT
$1.00
/ 1M tok
OUTPUT
$5.00
/ 1M tok
CONTEXT
200K
tokens
Compare with peers→

Forecast

Projected spend · single request
$0.00
Input volume
0
tokens
Prompt · 0 tok$0.00
Response · ~0 tok projected
$0.00
Monthly spend projection
1K req/mo
$0.00
10K req/mo
$0.00
100K req/mo
$0.00
Claude Haiku 4.5
200K context

Estimates only. Actual charges are set by the LLM provider at request time and may differ. Learn more.

Modelling a multi-turn workflow? Open the Session Simulator →

Keep exploring

  • Compare every model side-by-side →
  • Cost calculators for chatbots, RAG, agents →
  • Read the Q2 2026 pricing report →
  • Token counters for GPT, Claude, Gemini →
  • Unit converters (words ↔ tokens ↔ dollars) →
  • Full model pricing reference →