Menu

Tools

LLM Cost Calculator
Compare API costs across GPT-4, Claude, Deepseek, Gemini, and more. Find the most cost-effective AI model for your use case.
Usage Parameters
Configure your expected token usage to compare costs
1K1M
100500K
110K

Cheapest Option

Gemini 2.5 Flash

$67.50/month

Premium Model

Claude Opus 4.5

$15,750/month

Sort by:
Gemini 2.5 Flash

Google

Cheapest
Ultra Fast

Speed optimized, huge context

Input$0.075/M tokens
Output$0.3/M tokens
Context1000K tokens
Per Request$0.022
Daily$2.25
Monthly$67.50
Yearly$821
GPT-5 Nano

OpenAI

Cheapest

Ultra-low cost for simple tasks

Input$0.05/M tokens
Output$0.4/M tokens
Cached Input$0.005/M tokens
Context64K tokens
Per Request$0.025
Daily$2.50
Monthly$75.00
Yearly$913
DeepSeek V3.2

DeepSeek

Best Value

#1 open-source, exceptional coding

Input$0.28/M tokens
Output$0.42/M tokens
Cached Input$0.028/M tokens
Context64K tokens
Per Request$0.049
Daily$4.90
Monthly$147
Yearly$1,789
Mistral Medium 3

Mistral

Balanced

Great speed/cost balance

Input$0.4/M tokens
Output$1.2/M tokens
Context128K tokens
Per Request$0.100
Daily$10.00
Monthly$300
Yearly$3,650
GPT-5 Mini

OpenAI

Budget

Affordable & fast for everyday tasks

Input$0.25/M tokens
Output$2/M tokens
Cached Input$0.025/M tokens
Context128K tokens
Per Request$0.125
Daily$12.50
Monthly$375
Yearly$4,563
DeepSeek-R1

DeepSeek

Reasoning

Reasoning model rivaling o1

Input$0.55/M tokens
Output$2.19/M tokens
Cached Input$0.14/M tokens
Context64K tokens
Per Request$0.165
Daily$16.45
Monthly$494
Yearly$6,004
Claude Haiku 4

Anthropic

Fast

Fastest Claude for quick responses

Input$0.8/M tokens
Output$4/M tokens
Context200K tokens
Per Request$0.280
Daily$28.00
Monthly$840
Yearly$10,220
Mistral Large 3

Mistral

Flagship

675B params, top-tier performance

Input$2/M tokens
Output$6/M tokens
Context128K tokens
Per Request$0.500
Daily$50.00
Monthly$1,500
Yearly$18,250
GPT-5

OpenAI

Flagship

Latest flagship model, 400K context

Input$1.25/M tokens
Output$10/M tokens
Cached Input$0.125/M tokens
Context400K tokens
Per Request$0.625
Daily$62.50
Monthly$1,875
Yearly$22,813
Gemini 2.5 Pro

Google

1M Context

Best for complex reasoning & code

Input$1.25/M tokens
Output$10/M tokens
Context1000K tokens
Per Request$0.625
Daily$62.50
Monthly$1,875
Yearly$22,813
GPT-4o

OpenAI

Previous gen multimodal model

Input$2.5/M tokens
Output$10/M tokens
Cached Input$1.25/M tokens
Context128K tokens
Per Request$0.750
Daily$75.00
Monthly$2,250
Yearly$27,375
Gemini 3 Pro (Preview)

Google

Preview

Latest multimodal, 2M context

Input$2/M tokens
Output$12/M tokens
Context2000K tokens
Per Request$0.800
Daily$80.00
Monthly$2,400
Yearly$29,200
Claude Sonnet 4.5

Anthropic

Popular

Best for coding, agents & agentic tasks

Input$3/M tokens
Output$15/M tokens
Context200K tokens
Per Request$1.05
Daily$105
Monthly$3,150
Yearly$38,325
Claude Opus 4.5

Anthropic

Most Capable

Most intelligent Claude, infinite chats

Input$15/M tokens
Output$75/M tokens
Context200K tokens
Per Request$5.25
Daily$525
Monthly$15,750
Yearly$191,625
About LLM Pricing

💡 Tips to Reduce Costs

  • • Use prompt caching for repeated contexts
  • • Choose smaller models for simple tasks
  • • Optimize prompts to reduce token count
  • • Batch requests when possible

📊 Understanding Tokens

  • • ~4 characters = 1 token (English)
  • • ~750 words ≈ 1,000 tokens
  • • Output tokens cost more than input
  • • Context window limits total tokens

⚠️ Disclaimer: Prices are updated periodically and may not reflect the latest rates. Always check the official provider documentation for current pricing.

About LLM Cost Calculator

Compare pricing across OpenAI, Anthropic, Google, and other AI providers. Enter your expected usage and see cost estimates for GPT-4, Claude, Gemini, Deepseek, and more.

AI API costs add up quickly at scale. This calculator helps you compare models side-by-side, estimate monthly costs, and find the most cost-effective option for your use case without surprise bills.

How to use LLM Cost Calculator

1

Enter your expected input tokens per request.

2

Enter expected output tokens per request.

3

Set your estimated requests per day.

4

View cost comparison across all models.

5

Adjust parameters to explore different scenarios.

Examples

Chatbot cost estimation

Typical customer service chatbot scenario:

Input: 500 tokens (user message + context)
Output: 200 tokens (response)
Volume: 1,000 requests/day

GPT-4o:      $X/month
Claude 3.5:  $Y/month
Gemini Pro:  $Z/month

Choose based on quality vs. cost tradeoff.

Document processing

Processing long documents for summarization:

Input: 10,000 tokens (full document)
Output: 500 tokens (summary)
Volume: 100 docs/day

Long context = high input costs.
Consider: is 128k context worth the premium
over chunking with smaller context models?

Batch vs. real-time

Background processing can save 50%:

Real-time API: Full price, instant results
Batch API:     50% discount, hours delay

For non-urgent tasks (reports, analysis),
batch processing cuts costs in half.

Features

Compare 20+ AI models from major providers
Input/output token pricing breakdown
Per-request, daily, and monthly estimates
Real-time pricing data (updated regularly)
Batch vs. real-time pricing comparison
Share calculation links with teams

When to use this

  • Budgeting for AI-powered product features
  • Comparing models for cost optimization
  • Estimating startup runway with AI costs
  • Choosing between GPT-4 vs Claude vs Gemini
  • Planning batch processing vs real-time API usage
  • Justifying AI investments to stakeholders

Common questions