Free · No signup · 100% client-side
AI Token Cost Calculator
Estimate AI API costs for OpenAI, Claude, Gemini, DeepSeek and Mistral in seconds. Enter tokens, compare models side by side, and avoid surprise bills.
Optional: estimate tokens from sample text
Estimated total cost
$0.00
- Per request
- $0.00
- Input cost
- $0.00
- Cached input cost
- $0.00
- Output cost
- $0.00
Total tokens (in+out): 0· Requests: 0
Compare 10 cheapest models (same usage)
| Provider | Model | Per request | Total |
|---|
Cheapest option is highlighted. Prices are USD list rates for planning.
Static snapshot · fetching live catalog…
100% client-side · Data never leaves your device · Free forever
How to use it
- Pick a model from OpenAI, Anthropic (Claude), Google (Gemini), DeepSeek, or Mistral.
- Enter input and output tokens, plus cached input and number of requests.
- Read the total cost and compare the same usage across every model.
How LLM API pricing works
LLM providers bill by tokens, small chunks of text a model processes. Pricing is quoted per 1 million tokens, with a lower rate for input and a higher rate for output (generated) tokens. Rule of thumb: one token is roughly 4 characters of English text, about 0.75 words.
Reference pricing (USD per 1M tokens)
| Provider | Model | Input / 1M | Output / 1M | Cached in / 1M |
|---|---|---|---|---|
| OpenAI | GPT-5 | $1.25 | $10 | $0.125 |
| OpenAI | GPT-5 mini | $0.25 | $2 | $0.025 |
| OpenAI | GPT-5 nano | $0.05 | $0.4 | $0.005 |
| OpenAI | GPT-5.1 | $1.25 | $10 | $0.125 |
| Anthropic | Claude Sonnet 5 | $2 | $10 | $0.2 |
| Anthropic | Claude Opus 4.8 | $5 | $25 | $0.5 |
| Anthropic | Claude Haiku 4.5 | $1 | $5 | $0.1 |
| Gemini 2.5 Pro | $1.25 | $10 | $0.125 | |
| Gemini 2.5 Flash | $0.3 | $2.5 | $0.03 | |
| Gemini 3.5 Flash | $1.5 | $9 | $0.15 | |
| Gemini 3.6 Flash | $0.75 | $3.75 | $0.075 | |
| DeepSeek | DeepSeek V3 | $0.32 | $0.89 | |
| DeepSeek | DeepSeek R1 | $0.5 | $2.15 | $0.35 |
| DeepSeek | DeepSeek V4 Pro | $0.57948 | $1.73844 | $0.057948 |
| DeepSeek | DeepSeek V4 Flash | $0.065 | $0.18 | $0.016 |
| Mistral | Mistral Large 3 | $0.5 | $1.5 | $0.05 |
| Mistral | Mistral Medium 3 | $0.4 | $2 | $0.04 |
| xAI | Grok 4.3 | $1.25 | $2.5 | $0.2 |
| Qwen | Qwen3 Max | $0.78 | $3.9 | $0.156 |
| MoonshotAI | Kimi K2.5 | $0.45 | $2.25 | $0.07 |
List prices for planning. Providers change pricing, so verify before large commits.
Frequently asked questions
Is the AI token cost calculator free?
Yes. The AI token cost calculator is completely free, runs 100% in your browser, and requires no account or signup.
How accurate are the AI API prices?
The calculator uses the latest published list prices for each model per 1M tokens. Prices are for planning and may not include discounts, bulk rates, or provider promotions.
Do you support OpenAI, Anthropic, and Google models?
Yes. The live calculator searches the full OpenRouter model catalog (300+ models across OpenAI, Anthropic, Google, DeepSeek, Mistral, xAI, Qwen, MoonshotAI, and more), with per-model input, output, and cached-input pricing refreshed from the public OpenRouter API.
How much does 300k tokens cost?
It depends on the model. For example, 300k input tokens on GPT-5 nano (~$0.05/1M) costs about $0.015, while on GPT-5 (~$1.25/1M) it costs about $0.375. Enter 300000 in the input field above and compare all models side by side.
How do I estimate Claude API costs?
Pick a Claude model above (e.g. Claude 3.5 Sonnet or Haiku), enter your input and output tokens plus request count. The calculator applies Anthropic's per-1M list prices and shows total, per-request, and cached-input savings instantly.
What is a token budget and how do I plan it?
A token budget is your planned input + output tokens multiplied by request volume. Estimate a single representative request above, set the number of requests (e.g. 1000), and use the total as your monthly budget baseline. The compare table shows the cheapest model for the same workload.
More free tools
Private, fast, and ready to use. No signup.
JSON Formatter & Validator
Format, validate, and minify JSON in one click.
Open tool →Regex Tester
Test regular expressions with instant match highlighting.
Open tool →Cron Expression Generator
Build cron schedules and preview exactly when they run.
Open tool →JWT Decoder
Read the contents of any JWT token safely in your browser.
Open tool →UUID Generator
Generate unique IDs in bulk for apps, databases, and tests.
Open tool →Base64 Encode & Decode
Encode or decode Base64 text in seconds.
Open tool →Hash Generator
Generate SHA-1, SHA-256, SHA-384, and SHA-512 hashes online.
Open tool →Open Graph & Meta Preview
Create Open Graph and social share tags for any page.
Open tool →Unix Timestamp Converter
Convert Unix timestamps to dates and timezones instantly.
Open tool →Planning Poker
Ephemeral scrum poker rooms, no login, no data saved.
Open tool →HTML to Markdown Converter
Convert HTML pages and snippets to clean Markdown instantly.
Open tool →SQL Formatter
Format and beautify SQL queries with proper indentation and uppercase keywords.
Open tool →CSV to JSON Converter
Convert CSV data to JSON and back, right in your browser.
Open tool →Password Generator & Password Strength Checker
Generate strong, secure random passwords and check their strength.
Open tool →JWT Encoder
Create a JWT token from header and payload, with optional signature.
Open tool →AI Prompt Builder
Assemble structured prompts with system, context, and task sections, and estimate token cost.
Open tool →Color Converter
Convert colors between HEX, RGB, and HSL with an instant preview.
Open tool →