A token is a unit of text processed by the API. On average: 1 token ≈ 4 characters in English, 1 token ≈ ¾ words, 100 tokens ≈ 75 words. The exact token count can vary depending on language and text type.

Which model should I choose?

The right model choice depends on your use case: Budget options: Gemini 1.5 Flash-8B ($0.0375/$0.15), Gemini 2.0 Flash-Lite ($0.075/$0.30), GPT-4.1-mini ($0.40/$1.60), Claude Haiku 3 ($0.25/$1.25). Best value: GPT-4.1-mini, Claude Haiku 3.5, Gemini 2.5 Flash-Lite. Premium models: Claude Opus 4, Gemini 2.5 Pro, GPT-4.5-preview. Special applications: O-series for reasoning, Audio/Realtime for real-time, Gemini 1.5 Pro for long documents.

Which API should I choose - OpenAI, Claude, or Gemini?

Each provider has its strengths: OpenAI: Largest model selection, established ecosystem, best tooling support. Anthropic Claude: Excellent safety, longer contexts (200K), great for complex tasks. Google Gemini: Very large context windows (up to 2M), affordable prices, strong multimodal features. Price comparison of cheapest models: Gemini 1.5 Flash-8B: $0.0375/$0.15, Gemini 2.0 Flash-Lite: $0.075/$0.30, GPT-4o-mini: $0.15/$0.60, Claude Haiku 3: $0.25/$1.25.

How can I save costs?

There are several ways to reduce API costs: Use shorter, more precise prompts. Use cheaper models like GPT-4o-mini for simple tasks. Implement caching for frequent requests. Set max_tokens limits for responses. Use batch processing for multiple requests.

What's the difference between input and output tokens?

Input tokens are all tokens you send to the API (your prompt, system message, etc.). Output tokens are the tokens in the API's response. Output tokens are typically more expensive because they need to be generated by the API, while input tokens are only processed.

Are there volume discounts, batch processing, or free tiers?

Each provider has different discount structures: OpenAI: New accounts often receive free credits, enterprise pricing available, Batch API offers up to 50% discount for delayed processing. Anthropic Claude: Batch processing with 50% discount available, prompt caching for repeated requests, enterprise contracts with volume discounts. Google Gemini: Free tier available for developers, context caching reduces costs for long contexts, lower prices for longer contexts (>128K/200K).

How accurate are these price calculations?

This calculator is based on official API prices from OpenAI, Anthropic, and Google (as of February 2026). The calculations are estimates because actual token counts depend on various factors: Token counting varies by model and tokenizer. Gemini prices may vary by context length. Batch processing and caching can reduce costs. Enterprise customers often receive individual pricing. For exact calculations, use the official tokenizers from each provider.

55+ AI Models

AI API Cost Calculator

Calculate costs for OpenAI, Anthropic Claude and Google Gemini APIs with all available models.

New: New: Claude 4.6, Gemini 3.1 Pro and more models now available! Compare 55+ different AI models.

Filter & Search

Find the perfect model for your requirements

53 models found

Calculation

Enter your parameters to calculate costs

Input format

Input Tokens

≈ 1000 Tokens

Output Tokens

≈ 1000 Tokens

Number of API calls

Select model

Total cost:$0.0187

Cost per call:$0.000188

Input cost:$0.0037

Output cost:$0.0150

Model Comparison

Cost comparison of all available AI models

Model	API Provider	Input $/1M	Output $/1M	Context	Total Cost
Gemini 1.5 Flash-8B Smallest model for simple tasks	Google	$0.04	$0.15	1M	$0.0187 Cheapest Option
Gemini 2.0 Flash-Lite Optimized for scaling	Google	$0.07	$0.30	1M	$0.0375
Gemini 1.5 Flash Fast multimodal model	Google	$0.07	$0.30	1M	$0.0375
GPT-5 Nano Most cost-effective GPT-5 model with impressive performance New	OpenAI	$0.05	$0.40	272K	$0.0450
Llama 4 Scout 109B parameter open-source model Open Source	Groq / Together AI	$0.11	$0.34	10M	$0.0450
Gemini 2.5 Flash-Lite Most cost-effective Gemini model	Google	$0.10	$0.40	1M	$0.0500
Gemini 2.0 Flash Balanced multimodal model	Google	$0.10	$0.40	1M	$0.0500
GPT-4o-mini Fastest and most cost-effective GPT-4o model Popular	OpenAI	$0.15	$0.60	128K	$0.0750
Grok 3 Mini Budget-friendly Grok model	xAI	$0.30	$0.50	131K	$0.0800
Mistral Small Compact and fast model	Mistral AI	$0.20	$0.60	128K	$0.0800
Llama 4 Maverick 400B parameter premium performance Open Source	Groq / Together AI	$0.50	$0.77	1M	$0.1270
Claude Haiku 3 Older Haiku model - very affordable	Anthropic	$0.25	$1.25	200K	$0.1500
GPT-4.1-mini Compact model with excellent price-performance ratio Popular	OpenAI	$0.40	$1.60	1M	$0.2000
GPT-3.5 Turbo Classic ChatGPT model Legacy	OpenAI	$0.50	$1.50	16K	$0.2000
GPT-5 Mini 80% of GPT-5 performance at 20% of the cost Recommended	OpenAI	$0.25	$2.00	272K	$0.2250
Gemini 2.5 Flash Hybrid reasoning with adaptive thinking	Google	$0.30	$2.50	1M	$0.2800
Gemini 3 Flash Frontier intelligence at Flash speed	Google	$0.60	$3.00	1M	$0.3600
Claude Haiku 3.5 Fastest and most affordable Claude model	Anthropic	$0.80	$4.00	200K	$0.4800
o1-mini Compact reasoning model Reasoning	OpenAI	$1.10	$4.40	128K	$0.5500
o3-mini Newest mini reasoning model Reasoning	OpenAI	$1.10	$4.40	200K	$0.5500
o4-mini Latest generation mini reasoning	OpenAI	$1.10	$4.40	200K	$0.5500
Gemini 1.5 Pro Highest intelligence with 2M token context	Google	$1.25	$5.00	2M	$0.6250
GPT-3.5 Turbo 16K GPT-3.5 with extended context Legacy	OpenAI	$3.00	$4.00	16K	$0.7000
Mistral Large 2 Flagship with 128K context	Mistral AI	$2.00	$6.00	128K	$0.8000
GPT-4.1 GPT-4.1 generation with extended context	OpenAI	$2.00	$8.00	1M	$1.0000
o3 Improved o3 with 80% price reduction Budget	OpenAI	$2.00	$8.00	200K	$1.0000
GPT-5 Flagship model with reasoning and multimodal capabilities Premium	OpenAI	$1.25	$10.00	272K	$1.1250
GPT-5 Chat Optimized for conversations and dialogues	OpenAI	$1.25	$10.00	272K	$1.1250
GPT-4o Optimized GPT-4 model Recommended	OpenAI	$2.50	$10.00	128K	$1.2500
GPT-4o Audio GPT-4o with audio processing (text) Audio	OpenAI	$2.50	$10.00	128K	$1.2500
Gemini 3.1 Pro 2x reasoning leap, GPQA Diamond record (94.3%) New	Google	$2.00	$12.00	1M	$1.4000
Gemini 3 Pro Frontier intelligence with deep research	Google	$2.00	$12.00	1M	$1.4000
Gemini 2.5 Pro State-of-the-art thinking model with Deep Think	Google	$2.50	$15.00	1M	$1.7500
Claude Sonnet 4.5 Sonnet with improved performance	Anthropic	$3.00	$15.00	200K	$1.8000
Claude Sonnet 4.6 Opus-level performance at Sonnet pricing, 1M context New	Anthropic	$3.00	$15.00	1M	$1.8000
Claude Sonnet 4 Optimal balance between intelligence and cost	Anthropic	$3.00	$15.00	200K	$1.8000
Claude Sonnet 3.7 Improved Sonnet model	Anthropic	$3.00	$15.00	200K	$1.8000
Claude Sonnet 3.5 Proven Sonnet model	Anthropic	$3.00	$15.00	200K	$1.8000
Grok 4 Latest flagship model with reasoning	xAI	$3.00	$15.00	256K	$1.8000
Grok 3 Standard Grok with 1M context	xAI	$3.00	$15.00	131K	$1.8000
GPT-4o Realtime Real-time processing with GPT-4o (text) Realtime	OpenAI	$5.00	$20.00	128K	$2.5000
Claude Opus 4.6 Flagship model with 1M context and agent teams New	Anthropic	$5.00	$25.00	1M	$3.0000
Grok 4 Heavy Most powerful Grok model (SuperGrok Heavy) Premium	xAI	$5.00	$25.00	256K	$3.0000
Grok 3 Speed Fastest Grok variant	xAI	$5.00	$25.00	131K	$3.0000
GPT-4 Turbo Faster GPT-4 variant Legacy	OpenAI	$10.00	$30.00	128K	$4.0000
o1 Advanced reasoning model Premium	OpenAI	$15.00	$60.00	200K	$7.5000
GPT-4 Original GPT-4 model Legacy	OpenAI	$30.00	$60.00	8K	$9.0000
Claude Opus 4.1 Opus model with 74.5% SWE-bench score	Anthropic	$15.00	$75.00	200K	$9.0000
Claude Opus 4.5 Extended thinking, highest intelligence	Anthropic	$15.00	$75.00	200K	$9.0000
Claude Opus 4 Intelligent Claude model	Anthropic	$15.00	$75.00	200K	$9.0000
Claude Opus 3 Previous generation with high intelligence	Anthropic	$15.00	$75.00	200K	$9.0000
o3-pro Premium o3 model with maximum performance Premium	OpenAI	$25.00	$100.00	200K	$12.5000
GPT-4-32K GPT-4 with extended context Legacy	OpenAI	$60.00	$120.00	32K	$18.0000

Note on Pricing

Displayed prices are based on official API pricing (as of February 2026). Audio and Realtime models show text-based prices only. Gemini prices may vary by context length (standard prices shown). Actual costs may vary based on usage, discounts, and specific features.

Frequently Asked Questions

Important information about using AI APIs

Token Conversion & Examples

Understand how tokens are counted

Average Conversions

1 Token≈ 4 characters (English)

1 Token≈ 2-3 characters (German)

1 Word≈ 1.3 tokens

1 Sentence (15 words)≈ 20 tokens

1 Paragraph (100 words)≈ 130 tokens

Typical Use Cases

Simple Q&A

Input: ~50 tokens, Output: ~100 tokens

Gemini 1.5 Flash-8B: $0.000017 | GPT-4o-mini: $0.000068 | Claude Haiku 3: $0.000138

Text Summary

Input: ~1000 tokens, Output: ~200 tokens

Gemini Flash-8B: $0.000068 | Gemini 2.0 Flash-Lite: $0.000135 | GPT-4o-mini: $0.000195

Code Generation

Input: ~200 tokens, Output: ~500 tokens

≈ $0.00031 with GPT-4o-mini | ≈ $0.00053 with Claude Haiku 3

Complex Reasoning

Input: ~500 tokens, Output: ~1000 tokens

≈ $0.00495 with O3 | ≈ $0.01875 with O1

Changelog

Die neuesten Updates und Verbesserungen unseres API Cost Calculator

v3.0.0February 20, 2026

New Models: Claude 4.6, Gemini 3 Series

• Added Claude Opus 4.6 with 1M context and agent teams
• Added Claude Sonnet 4.6 with Opus-level performance
• Added Claude Opus 4.5 with extended thinking
• Added Gemini 3.1 Pro with GPQA Diamond record
• Added Gemini 3 Pro and Gemini 3 Flash
• Updated badge system (removed stale New badges)
• Total of 55+ AI models available

v2.0.0September 15, 2025

Major Update with New Models

• Added GPT-5 series (Nano, Mini, Standard, Chat)
• Claude 4.1 Opus with 74.5% SWE-bench score
• Added Claude 4.5 Sonnet
• xAI Grok models (Grok 3 & 4 series)
• Meta Llama 4 models (Scout & Maverick)
• Added Mistral Large 2 and Small
• Updated Gemini 2.5 Pro pricing
• Total of 50+ AI models available

v1.2.1July 23, 2025

Changelog Added

• Added changelog section to API cost calculator

v1.2.0July 22, 2025

Major Update

• Added all Claude 4 and Gemini 2.5 models
• Implemented comprehensive filter and sort functions
• Extended comparison options for 36 AI models

v1.1.0July 22, 2025

New Models

• Added GPT-4.1 series with 1M token context
• Updated O3 model with 80% price reduction
• Integrated new audio and realtime model variants

v1.0.0July 22, 2025

Initial Release

• Initial version with 21 OpenAI models
• Token conversion for words and characters
• Interactive cost calculation and comparison table

Model

API Provider

Input $/1M

Output $/1M

Context

Total Cost

Gemini 1.5 Flash-8B

Smallest model for simple tasks

Google

$0.04

$0.15

$0.0187

Cheapest Option

Gemini 2.0 Flash-Lite

Optimized for scaling

Google

$0.07

$0.30

$0.0375

Gemini 1.5 Flash

Fast multimodal model

Google

$0.07

$0.30

$0.0375

GPT-5 Nano

Most cost-effective GPT-5 model with impressive performance

New

OpenAI

$0.05

$0.40

272K

$0.0450

Llama 4 Scout

109B parameter open-source model

Open Source

Groq / Together AI

$0.11

$0.34

10M

$0.0450

Gemini 2.5 Flash-Lite

Most cost-effective Gemini model

Google

$0.10

$0.40

$0.0500

Gemini 2.0 Flash

Balanced multimodal model

Google

$0.10

$0.40

$0.0500

GPT-4o-mini

Fastest and most cost-effective GPT-4o model

Popular

OpenAI

$0.15

$0.60

128K

$0.0750

Grok 3 Mini

Budget-friendly Grok model

xAI

$0.30

$0.50

131K

$0.0800

Mistral Small

Compact and fast model

Mistral AI

$0.20

$0.60

128K

$0.0800

Llama 4 Maverick

400B parameter premium performance

Open Source

Groq / Together AI

$0.50

$0.77

$0.1270

Claude Haiku 3

Older Haiku model - very affordable

Anthropic

$0.25

$1.25

200K

$0.1500

GPT-4.1-mini

Compact model with excellent price-performance ratio

Popular

OpenAI

$0.40

$1.60

$0.2000

GPT-3.5 Turbo

Classic ChatGPT model

Legacy

OpenAI

$0.50

$1.50

16K

$0.2000

GPT-5 Mini

80% of GPT-5 performance at 20% of the cost

Recommended

OpenAI

$0.25

$2.00

272K

$0.2250

Gemini 2.5 Flash

Hybrid reasoning with adaptive thinking

Google

$0.30

$2.50

$0.2800

Gemini 3 Flash

Frontier intelligence at Flash speed

Google

$0.60

$3.00

$0.3600

Claude Haiku 3.5

Fastest and most affordable Claude model

Anthropic

$0.80

$4.00

200K

$0.4800

o1-mini

Compact reasoning model

Reasoning

OpenAI

$1.10

$4.40

128K

$0.5500

o3-mini

Newest mini reasoning model

Reasoning

OpenAI

$1.10

$4.40

200K

$0.5500

o4-mini

Latest generation mini reasoning

OpenAI

$1.10

$4.40

200K

$0.5500

Gemini 1.5 Pro

Highest intelligence with 2M token context

Google

$1.25

$5.00

$0.6250

GPT-3.5 Turbo 16K

GPT-3.5 with extended context

Legacy

OpenAI

$3.00

$4.00

16K

$0.7000

Mistral Large 2

Flagship with 128K context

Mistral AI

$2.00

$6.00

128K

$0.8000

GPT-4.1

GPT-4.1 generation with extended context

OpenAI

$2.00

$8.00

$1.0000

Improved o3 with 80% price reduction

Budget

OpenAI

$2.00

$8.00

200K

$1.0000

GPT-5

Flagship model with reasoning and multimodal capabilities

Premium

OpenAI

$1.25

$10.00

272K

$1.1250

GPT-5 Chat

Optimized for conversations and dialogues

OpenAI

$1.25

$10.00

272K

$1.1250

GPT-4o

Optimized GPT-4 model

Recommended

OpenAI

$2.50

$10.00

128K

$1.2500

GPT-4o Audio

GPT-4o with audio processing (text)

Audio

OpenAI

$2.50

$10.00

128K

$1.2500

Gemini 3.1 Pro

2x reasoning leap, GPQA Diamond record (94.3%)

New

Google

$2.00

$12.00

$1.4000

Gemini 3 Pro

Frontier intelligence with deep research

Google

$2.00

$12.00

$1.4000

Gemini 2.5 Pro

State-of-the-art thinking model with Deep Think

Google

$2.50

$15.00

$1.7500

Claude Sonnet 4.5

Sonnet with improved performance

Anthropic

$3.00

$15.00

200K

$1.8000

Claude Sonnet 4.6

Opus-level performance at Sonnet pricing, 1M context

New

Anthropic

$3.00

$15.00

$1.8000

Claude Sonnet 4

Optimal balance between intelligence and cost

Anthropic

$3.00

$15.00

200K

$1.8000

Claude Sonnet 3.7

Improved Sonnet model

Anthropic

$3.00

$15.00

200K

$1.8000

Claude Sonnet 3.5

Proven Sonnet model

Anthropic

$3.00

$15.00

200K

$1.8000

Grok 4

Latest flagship model with reasoning

xAI

$3.00

$15.00

256K

$1.8000

Grok 3

Standard Grok with 1M context

xAI

$3.00

$15.00

131K

$1.8000

GPT-4o Realtime

Real-time processing with GPT-4o (text)

Realtime

OpenAI

$5.00

$20.00

128K

$2.5000

Claude Opus 4.6

Flagship model with 1M context and agent teams

New

Anthropic

$5.00

$25.00

$3.0000

Grok 4 Heavy

Most powerful Grok model (SuperGrok Heavy)

Premium

xAI

$5.00

$25.00

256K

$3.0000

Grok 3 Speed

Fastest Grok variant

xAI

$5.00

$25.00

131K

$3.0000

GPT-4 Turbo

Faster GPT-4 variant

Legacy

OpenAI

$10.00

$30.00

128K

$4.0000

Advanced reasoning model

Premium

OpenAI

$15.00

$60.00

200K

$7.5000

GPT-4

Original GPT-4 model

Legacy

OpenAI

$30.00

$60.00

$9.0000

Claude Opus 4.1

Opus model with 74.5% SWE-bench score

Anthropic

$15.00

$75.00

200K

$9.0000

Claude Opus 4.5

Extended thinking, highest intelligence

Anthropic

$15.00

$75.00

200K

$9.0000

Claude Opus 4

Intelligent Claude model

Anthropic

$15.00

$75.00

200K

$9.0000

Claude Opus 3

Previous generation with high intelligence

Anthropic

$15.00

$75.00

200K

$9.0000

o3-pro

Premium o3 model with maximum performance

Premium

OpenAI

$25.00

$100.00

200K

$12.5000

GPT-4-32K

GPT-4 with extended context

Legacy

OpenAI

$60.00

$120.00

32K

$18.0000

AI API Cost Calculator

Note on Pricing

Frequently Asked Questions

What is a token?

Which model should I choose?

Which API should I choose - OpenAI, Claude, or Gemini?

How can I save costs?

What's the difference between input and output tokens?

Are there volume discounts, batch processing, or free tiers?

How accurate are these price calculations?

Average Conversions

Typical Use Cases

New Models: Claude 4.6, Gemini 3 Series

Major Update with New Models

Changelog Added

Major Update

New Models

Initial Release

AI API Cost Calculator

Note on Pricing

Frequently Asked Questions

What is a token?

Which model should I choose?

Which API should I choose - OpenAI, Claude, or Gemini?

How can I save costs?

What's the difference between input and output tokens?

Are there volume discounts, batch processing, or free tiers?

How accurate are these price calculations?

Average Conversions

Typical Use Cases

New Models: Claude 4.6, Gemini 3 Series

Major Update with New Models

Changelog Added

Major Update

New Models

Initial Release