How is LLM API pricing calculated?

LLM APIs charge per token, priced per million tokens ($/1M). Input and output have separate rates — output is typically 3-5x more expensive. The total cost = (input tokens / 1M × input rate) + (output tokens / 1M × output rate).

What is prompt caching and how does it save money?

Prompt caching lets you reuse previously processed input tokens at a reduced rate (typically 90% cheaper). For Anthropic, cached reads cost $0.30/1M vs $3/1M standard for Claude Sonnet. This is ideal for system prompts or repeated context.

Which LLM is the cheapest for production use?

It depends on your quality requirements. For general tasks: DeepSeek V3 ($0.27/$1.10 per 1M) and Gemini 2.0 Flash Lite ($0.075/$0.30) are extremely cost-effective. For high-quality outputs: Claude Haiku 4.5 ($1/$5) and GPT-4.1 mini ($0.40/$1.60) offer excellent quality-to-cost ratios.

How can I reduce my LLM API costs?

Three proven approaches: use prompt caching for repeated system prompts (saves up to 90%), switch to a smaller model for simple tasks (GPT-4.1 Mini or Claude Haiku), and use batch API for non-real-time workloads (typically 50% cheaper).

Why do input and output tokens have different prices?

Generation is computationally heavier. Output tokens require autoregressive decoding — one token at a time — while input tokens are processed in parallel. Most providers charge output at 2× to 5× the input rate.

Is the pricing data updated automatically?

Yes. Pricing syncs daily from upstream provider data via GitHub Actions. The timestamp in the toolbar shows the last sync. You can also click Refresh for a live update.

LLM Pricing

Compare LLM API costs across all major models

LLM Cost Calculator

64 models · 105 entries

Checked 6h ago

Input Tokens

tokens

Output Tokens

tokens

Calls / Day

/day

Quick estimates

Mistral Small 3CheapestValue

Mistral · mistral

$0.06

$0.18

$0.0001

$0.45

/mo

131k

Qwen TurboValue

Alibaba · qwen

$0.05

$0.2

$0.0002

$0.45

/mo

129k

GPT-5 Nano

OpenAI · openai

$0.05

$0.4

$0.0003

$0.75

/mo

272k

Llama 4 Scout

Meta · groq

$0.11

$0.34

$0.0003

$0.84

/mo

131k

Gemini 2.5 Flash Lite

Google · google

$0.1

$0.4

$0.0003

$0.90

/mo

GPT-4.1 Nano

OpenAI · openai

$0.1

$0.4

$0.0003

$0.90

/mo

GPT-4o Mini

OpenAI · openai

$0.15

$0.6

$0.0004

/mo

128k

DeepSeek V3.2Value

DeepSeek · deepseek

$0.28

$0.4

$0.0005

/mo

164k

DeepSeek Chat+1

DeepSeek · deepseek

$0.28

$0.42

$0.0005

/mo

131k

Llama 4 Maverick

Meta · groq

$0.2

$0.6

$0.0005

/mo

131k

Codestral

Mistral · mistral

$0.3

$0.9

$0.0008

/mo

256k

DeepSeek V3

DeepSeek · deepseek

$0.27

$1.1

$0.0008

/mo

66k

GPT-5.4 Nano

OpenAI · openai

$0.2

$1.25

$0.0008

/mo

272k

MiniMax M2.5+1

MiniMax · minimax

$0.3

$1.2

$0.0009

/mo

Llama 3.3 70B

Meta · groq

$0.59

$0.79

$0.0010

/mo

128k

Qwen Plus

Alibaba · qwen

$0.4

$1.2

$0.0010

/mo

129k

Gemini 3.1 Flash Lite

Google · google

$0.25

$1.5

$0.0010

/mo

Qwen Coder

Alibaba · qwen

$0.3

$1.5

$0.0010

/mo

GLM-4.7

Zhipu AI · openrouter

$0.4

$1.5

$0.0011

/mo

203k

GPT-4.1 Mini

OpenAI · openai

$0.4

$1.6

$0.0012

/mo

GPT-5 Mini

OpenAI · openai

$0.25

$0.0013

/mo

272k

Mistral Large 3

Mistral · mistral

$0.5

$1.5

$0.0013

/mo

262k

Mistral Medium 3

Mistral · mistral

$0.4

$0.0014

/mo

131k

Gemini 3.5 Flash+1

Google · google

$0.3

$2.5

$0.0015

/mo

Gemini 3 Flash

Google · google

$0.5

$0.0020

/mo

GLM-5

Zhipu AI · openrouter

$0.8

$2.56

$0.0021

/mo

203k

Kimi K2 Thinking

Moonshot AI · bedrock

$0.73

$3.03

$0.0022

/mo

262k

Grok 4.3

xAI · xai

$1.25

$2.5

$0.0025

/mo

Kimi K2.5

Moonshot AI · bedrock

$0.72

$3.6

$0.0025

/mo

262k

Kimi K2.6

Moonshot AI · moonshot

$0.95

$0.0029

/mo

262k

GPT-5.4 Mini

OpenAI · openai

$0.75

$4.5

$0.0030

/mo

272k

o3 Mini+1

OpenAI · openai

$1.1

$4.4

$0.0033

$10

/mo

200k

Claude Haiku 4.5

Anthropic · anthropic

$0.0035

$11

/mo

200k

GLM-5 Code

Zhipu AI · zai

$1.2

$0.0037

$11

/mo

200k

Qwen Max

Alibaba · qwen

$1.6

$6.4

$0.0048

$14

/mo

31k

Pixtral Large

Mistral · mistral

$0.0050

$15

/mo

128k

GPT-4.1

OpenAI · openai

$0.0060

$18

/mo

OpenAI · openai

$0.0060

$18

/mo

200k

Gemini 2.5 Pro

Google · google

$1.25

$10

$0.0063

$19

/mo

GPT-5+1

OpenAI · openai

$1.25

$10

$0.0063

$19

/mo

272k

Grok 2

xAI · xai

$10

$0.0070

$21

/mo

131k

GPT-4o

OpenAI · openai

$2.5

$10

$0.0075

$23

/mo

128k

Gemini 3 Pro+1

Google · vertex_ai

$12

$0.0080

$24

/mo

GPT-5.2

OpenAI · openai

$1.75

$14

$0.0088

$26

/mo

272k

GPT-5.4

OpenAI · openai

$2.5

$15

$0.010

$30

/mo

Claude Sonnet 4.6+1

Anthropic · anthropic

$15

$0.010

$31

/mo

Claude Sonnet 4.5

Anthropic · anthropic

$15

$0.010

$31

/mo

200k

Grok 4

xAI · xai

$15

$0.010

$31

/mo

256k

Claude 3.7 Sonnet

Anthropic · bedrock

$3.6

$18

$0.013

$38

/mo

200k

Claude Opus 4.5

Anthropic · anthropic

$25

$0.018

$53

/mo

200k

Claude Opus 4.6+2

Anthropic · anthropic

$25

$0.018

$53

/mo

GPT-5.5

OpenAI · openai

$30

$0.020

$60

/mo

Claude Fable 5

Anthropic · anthropic

$10

$50

$0.035

$105

/mo

OpenAI · openai

$15

$60

$0.045

$135

/mo

200k

Claude Opus 4

Anthropic · anthropic

$15

$75

$0.052

$158

/mo

200k

Mistral Small 3Best

Mistral · mistral

$0.0001

per request

Input

$0.06/M

Output

$0.18/M

Monthly

$0.45

Qwen Turbo

Alibaba · qwen

$0.0002

per request

Input

$0.05/M

Output

$0.2/M

Monthly

$0.45

GPT-5 Nano

OpenAI · openai

$0.0003

per request

Input

$0.05/M

Output

$0.4/M

Monthly

$0.75

Llama 4 Scout

Meta · groq

$0.0003

per request

Input

$0.11/M

Output

$0.34/M

Monthly

$0.84

Gemini 2.5 Flash Lite

Google · google

$0.0003

per request

Input

$0.1/M

Output

$0.4/M

Monthly

$0.90

GPT-4.1 Nano

OpenAI · openai

$0.0003

per request

Input

$0.1/M

Output

$0.4/M

Monthly

$0.90

GPT-4o Mini

OpenAI · openai

$0.0004

per request

Input

$0.15/M

Output

$0.6/M

Monthly

DeepSeek V3.2

DeepSeek · deepseek

$0.0005

per request

Input

$0.28/M

Output

$0.4/M

Monthly

DeepSeek Chat+1

DeepSeek · deepseek

$0.0005

per request

Input

$0.28/M

Output

$0.42/M

Monthly

Llama 4 Maverick

Meta · groq

$0.0005

per request

Input

$0.2/M

Output

$0.6/M

Monthly

Codestral

Mistral · mistral

$0.0008

per request

Input

$0.3/M

Output

$0.9/M

Monthly

DeepSeek V3

DeepSeek · deepseek

$0.0008

per request

Input

$0.27/M

Output

$1.1/M

Monthly

GPT-5.4 Nano

OpenAI · openai

$0.0008

per request

Input

$0.2/M

Output

$1.25/M

Monthly

MiniMax M2.5+1

MiniMax · minimax

$0.0009

per request

Input

$0.3/M

Output

$1.2/M

Monthly

Llama 3.3 70B

Meta · groq

$0.0010

per request

Input

$0.59/M

Output

$0.79/M

Monthly

Qwen Plus

Alibaba · qwen

$0.0010

per request

Input

$0.4/M

Output

$1.2/M

Monthly

Gemini 3.1 Flash Lite

Google · google

$0.0010

per request

Input

$0.25/M

Output

$1.5/M

Monthly

Qwen Coder

Alibaba · qwen

$0.0010

per request

Input

$0.3/M

Output

$1.5/M

Monthly

GLM-4.7

Zhipu AI · openrouter

$0.0011

per request

Input

$0.4/M

Output

$1.5/M

Monthly

GPT-4.1 Mini

OpenAI · openai

$0.0012

per request

Input

$0.4/M

Output

$1.6/M

Monthly

GPT-5 Mini

OpenAI · openai

$0.0013

per request

Input

$0.25/M

Output

$2/M

Monthly

Mistral Large 3

Mistral · mistral

$0.0013

per request

Input

$0.5/M

Output

$1.5/M

Monthly

Mistral Medium 3

Mistral · mistral

$0.0014

per request

Input

$0.4/M

Output

$2/M

Monthly

Gemini 3.5 Flash+1

Google · google

$0.0015

per request

Input

$0.3/M

Output

$2.5/M

Monthly

Gemini 3 Flash

Google · google

$0.0020

per request

Input

$0.5/M

Output

$3/M

Monthly

GLM-5

Zhipu AI · openrouter

$0.0021

per request

Input

$0.8/M

Output

$2.56/M

Monthly

Kimi K2 Thinking

Moonshot AI · bedrock

$0.0022

per request

Input

$0.73/M

Output

$3.03/M

Monthly

Grok 4.3

xAI · xai

$0.0025

per request

Input

$1.25/M

Output

$2.5/M

Monthly

Kimi K2.5

Moonshot AI · bedrock

$0.0025

per request

Input

$0.72/M

Output

$3.6/M

Monthly

Kimi K2.6

Moonshot AI · moonshot

$0.0029

per request

Input

$0.95/M

Output

$4/M

Monthly

GPT-5.4 Mini

OpenAI · openai

$0.0030

per request

Input

$0.75/M

Output

$4.5/M

Monthly

o3 Mini+1

OpenAI · openai

$0.0033

per request

Input

$1.1/M

Output

$4.4/M

Monthly

$10

Claude Haiku 4.5

Anthropic · anthropic

$0.0035

per request

Input

$1/M

Output

$5/M

Monthly

$11

GLM-5 Code

Zhipu AI · zai

$0.0037

per request

Input

$1.2/M

Output

$5/M

Monthly

$11

Qwen Max

Alibaba · qwen

$0.0048

per request

Input

$1.6/M

Output

$6.4/M

Monthly

$14

Pixtral Large

Mistral · mistral

$0.0050

per request

Input

$2/M

Output

$6/M

Monthly

$15

GPT-4.1

OpenAI · openai

$0.0060

per request

Input

$2/M

Output

$8/M

Monthly

$18

OpenAI · openai

$0.0060

per request

Input

$2/M

Output

$8/M

Monthly

$18

Gemini 2.5 Pro

Google · google

$0.0063

per request

Input

$1.25/M

Output

$10/M

Monthly

$19

GPT-5+1

OpenAI · openai

$0.0063

per request

Input

$1.25/M

Output

$10/M

Monthly

$19

Grok 2

xAI · xai

$0.0070

per request

Input

$2/M

Output

$10/M

Monthly

$21

GPT-4o

OpenAI · openai

$0.0075

per request

Input

$2.5/M

Output

$10/M

Monthly

$23

Gemini 3 Pro+1

Google · vertex_ai

$0.0080

per request

Input

$2/M

Output

$12/M

Monthly

$24

GPT-5.2

OpenAI · openai

$0.0088

per request

Input

$1.75/M

Output

$14/M

Monthly

$26

GPT-5.4

OpenAI · openai

$0.010

per request

Input

$2.5/M

Output

$15/M

Monthly

$30

Claude Sonnet 4.6+1

Anthropic · anthropic

$0.010

per request

Input

$3/M

Output

$15/M

Monthly

$31

Claude Sonnet 4.5

Anthropic · anthropic

$0.010

per request

Input

$3/M

Output

$15/M

Monthly

$31

Grok 4

xAI · xai

$0.010

per request

Input

$3/M

Output

$15/M

Monthly

$31

Claude 3.7 Sonnet

Anthropic · bedrock

$0.013

per request

Input

$3.6/M

Output

$18/M

Monthly

$38

Claude Opus 4.5

Anthropic · anthropic

$0.018

per request

Input

$5/M

Output

$25/M

Monthly

$53

Claude Opus 4.6+2

Anthropic · anthropic

$0.018

per request

Input

$5/M

Output

$25/M

Monthly

$53

GPT-5.5

OpenAI · openai

$0.020

per request

Input

$5/M

Output

$30/M

Monthly

$60

Claude Fable 5

Anthropic · anthropic

$0.035

per request

Input

$10/M

Output

$50/M

Monthly

$105

OpenAI · openai

$0.045

per request

Input

$15/M

Output

$60/M

Monthly

$135

Claude Opus 4

Anthropic · anthropic

$0.052

per request

Input

$15/M

Output

$75/M

Monthly

$158

The cheapest option (Mistral Small 3) saves $157/mo vs Claude Opus 4 (99.7% reduction)

Compare LLM API pricing across 64+ model families. Data synced daily from upstream providers. Supports standard, cached, and batch pricing modes.

Last checked: 7/31/2026

55 models shown