LLM Cost Calculator
Compare the cost of a prompt and response across models and providers. Enter token counts (or paste text to estimate them), pick a currency, and this tool estimates the cost using each provider's published pricing.
Input Tokens
0
Output Tokens
1,000
Requests
1
| Model | Input Cost | Output Cost | Total Cost | Comparison |
|---|---|---|---|---|
1 Gemini 1.5 Flash | $0 | $0.0003 | $0.0003 | Lowest |
2 Gemini 2.0 Flash | $0 | $0.0004 | $0.0004 | +33% |
3 GPT-4o Mini OpenAI | $0 | $0.0006 | $0.0006 | +100% |
GPT-4o Mini Azure | $0 | $0.0006 | $0.0006 | +100% |
Claude 3 Haiku Anthropic | $0 | $0.0013 | $0.0013 | +317% |
Claude 3 Haiku AWS Bedrock | $0 | $0.0013 | $0.0013 | +317% |
GPT-3.5 Turbo OpenAI | $0 | $0.0015 | $0.0015 | +400% |
Claude 3.5 Haiku Anthropic | $0 | $0.0040 | $0.0040 | +1233% |
o3 Mini OpenAI | $0 | $0.0045 | $0.0045 | +1407% |
Gemini 1.5 Pro | $0 | $0.0050 | $0.0050 | +1567% |
GPT-4o OpenAI | $0 | $0.010 | $0.010 | +3233% |
GPT-4o Azure | $0 | $0.010 | $0.010 | +3233% |
o1 Mini OpenAI | $0 | $0.012 | $0.012 | +3900% |
Claude Sonnet 4 Anthropic | $0 | $0.015 | $0.015 | +4900% |
Claude 3.5 Sonnet Anthropic | $0 | $0.015 | $0.015 | +4900% |
Claude 3.5 Sonnet AWS Bedrock | $0 | $0.015 | $0.015 | +4900% |
GPT-4 Turbo OpenAI | $0 | $0.030 | $0.030 | +9900% |
GPT-4 Turbo Azure | $0 | $0.030 | $0.030 | +9900% |
GPT-4 OpenAI | $0 | $0.060 | $0.060 | +19900% |
o1 OpenAI | $0 | $0.060 | $0.060 | +19900% |
GPT-4 Azure | $0 | $0.060 | $0.060 | +19900% |
Claude Opus 4.5 Anthropic | $0 | $0.075 | $0.075 | +24900% |
Claude 3 Opus Anthropic | $0 | $0.075 | $0.075 | +24900% |
Claude 3 Opus AWS Bedrock | $0 | $0.075 | $0.075 | +24900% |
Notes
- • Prices as of January 2025. Check official sites for current pricing.
- • Token estimation: ~4 English chars per token, ~2 Korean chars per token
- • Korean uses more tokens than English (~1.5-2x).
- • Exchange rates are auto-fetched from ExchangeRate-APIon page load.
Token Calculation Reference
- • OpenAI Tokenizer - Tokenizer for GPT models
- • Anthropic Token Counting - Token counting API for Claude models
- • Gemini Token Guide - Token guide for Gemini models
Pricing Sources
Check the official pricing pages below for latest information:
How to use
- Choose one or more models to compare.
- Enter input and output token counts, or paste text to estimate them.
- Read the estimated cost per model in your chosen currency.
Frequently asked questions
- How are tokens counted from pasted text?
- The tool estimates token counts from the text length. For exact billing, use the token count your provider's API or dashboard reports.
- Does this call any external service?
- Only when you pick a currency other than USD, to fetch a current exchange rate. Token counting and cost math happen locally.
- Are the prices always current?
- No. Providers change pricing over time, so treat this as an estimate and confirm against the provider's current rate card.
Privacy
Token counts and calculations run in your browser. Choosing a non-USD currency fetches a live exchange rate from a third-party API, which is the only network request this tool makes.