Free Tools/AI API Cost Calculator
1Calculate costs
2Optimize with Svivva

AI API Cost Calculator

Compare real-time pricing across GPT-4o, Claude 3.5, Gemini, Llama and more. Free, no signup.

GPT-4oClaude 3.5GeminiLlama14 models

Configure Your API Usage

10010K100K

Your system prompt + user message. ~750 words ≈ 1,000 tokens.

504K32K

AI response length. Short answer ≈ 100–300 tokens.

1010K1M

Total API calls across all users/endpoints.

Cheapest/month

$6.75

Gemini 1.5 Flash

Median/month

$225

GPT-4o

Most expensive

$1575

Claude 3 Opus

Cost Comparison — 14 Models

Sort by:
ModelPer callDailyMonthly
Cheapest

Gemini 1.5 Flash

Google
$0.000225$0.225$6.75

Gemini 2.0 Flash

Google
$0.000225$0.225$6.75

Mixtral 8x7B

Groq
$0.000360$0.360$10.80

GPT-4o mini

OpenAI
$0.000450$0.450$13.50

Llama 3.1 70B

Groq
$0.000985$0.985$29.55

Llama 3.3 70B

Groq
$0.000985$0.985$29.55

Claude 3.5 Haiku

Anthropic
$0.0028$2.80$84.00

GPT-4o

OpenAI
$0.0075$7.50$225

Gemini 1.5 Pro

Google
$0.0088$8.75$263

o1-mini

OpenAI
$0.0090$9.00$270

Claude 3.5 Sonnet

Anthropic
$0.010$10.50$315

GPT-4 Turbo

OpenAI
$0.025$25.00$750

o1

OpenAI
$0.045$45.00$1350

Claude 3 Opus

Anthropic
$0.052$52.50$1575

Prices updated Q1 2026. Always check official provider pricing pages for the latest rates.

Step 1 complete · Next →

You could save $2.70–$5.40/month with auto-routing

At your current usage, Svivva's model router picks the cheapest model that meets your quality bar — cutting costs 40–80% without changing your prompts. Cheapest option right now: Gemini 1.5 Flash at $6.75/mo.

Start saving — free

Frequently Asked Questions

How are AI API costs calculated?

AI APIs charge per token — roughly 4 characters or ¾ of a word. You're billed separately for input tokens (your prompt) and output tokens (the AI's response). This calculator multiplies your token counts by the per-million-token rate for each model.

Which model gives the best price/performance ratio?

GPT-4o mini, Claude 3.5 Haiku, and Gemini 1.5 Flash are generally the best value for most tasks. They're 10-50x cheaper than their premium counterparts with 80-90% of the quality for standard tasks.

How accurate is this calculator?

Very accurate for planning purposes. Prices are sourced directly from provider pricing pages. Real usage may vary slightly if your token counts differ from estimates, or if providers run promotions.

What's a typical token count for a chatbot message?

A short user message is 50-150 tokens. A detailed system prompt is 200-1,000 tokens. A typical API response is 100-500 tokens. Combined, a single chat turn costs roughly 300-1,500 tokens.

Can I reduce my AI API costs automatically?

Yes — tools like Svivva can route requests to the cheapest model that meets a quality threshold, automatically falling back to premium models only when needed.