AI API cost calculator
What a workload costs per month on each major model, from the providers’ own price lists.
| Model | Per request | Per month |
|---|---|---|
| Claude Haiku 5.5 Anthropic | $0.00045 | $45.00 |
| GPT-6 Luna OpenAI | $0.00045 | $45.00 |
| DeepSeek Flash DeepSeekPeak-hour price. Off-peak hours cost half. | $0.0012 | $120.00 |
| Gemini 3.5 Flash-Lite Google | $0.0019 | $185.00 |
| Gemini 3.8 Flash Google | $0.0034 | $337.50 |
| GPT-6.1 Terra OpenAI | $0.0045 | $450.00 |
| DeepSeek V4 Pro DeepSeekPeak-hour price. Off-peak hours cost half. | $0.0046 | $462.00 |
| Grok 4.7 xAI | $0.007 | $700.00 |
| Claude Sonnet 5.5 Anthropic | $0.009 | $900.00 |
| GPT-6.1 Sol OpenAI | $0.009 | $900.00 |
| Gemini 4 Argon Google | $0.009 | $900.00 |
| Claude Opus 5.5 Anthropic | $0.018 | $1,800.00 |
| Claude Fable 5.1 Anthropic | $0.045 | $4,500.00 |
| GPT-6 Astra OpenAI | $0.045 | $4,500.00 |
Prices per million tokens
| Model | Input | Cached input | Output | Notes | Source |
|---|---|---|---|---|---|
| Claude Fable 5.1 (Anthropic) | $10.00 | $0.25 | $50.00 | Batch: 50% off | docs.claude.com |
| Claude Opus 5.5 (Anthropic) | $4.00 | $0.20 | $20.00 | Batch: 50% off | docs.claude.com |
| Claude Sonnet 5.5 (Anthropic) | $2.00 | $0.10 | $10.00 | Batch: 50% off | docs.claude.com |
| Claude Haiku 5.5 (Anthropic) | $0.10 | $0.01 | $0.50 | Prompts over 100K tokens: $0.50 in, $2.50 out. Batch: 50% off | docs.claude.com |
| GPT-6 Astra (OpenAI) | $10.00 | $1.00 | $50.00 | Prompts over 272K tokens: $20.00 in, $75.00 out. Batch: 50% off | platform.openai.com |
| GPT-6.1 Sol (OpenAI) | $2.00 | $0.10 | $10.00 | Prompts over 272K tokens: $4.00 in, $15.00 out. Batch: 50% off | platform.openai.com |
| GPT-6.1 Terra (OpenAI) | $1.00 | $0.05 | $5.00 | Prompts over 272K tokens: $2.00 in, $7.50 out. Batch: 50% off | platform.openai.com |
| GPT-6 Luna (OpenAI) | $0.10 | $0.01 | $0.50 | Prompts over 272K tokens: $0.20 in, $0.75 out. Batch: 50% off | platform.openai.com |
| Gemini 4 Argon (Google) | $2.00 | $0.10 | $10.00 | Rises to $4.00 in, $20.00 out on December 1, 2026. Batch: 50% off | ai.google.dev |
| Gemini 3.8 Flash (Google) | $0.75 | $0.075 | $3.75 | Rises to $1.50 in, $7.50 out on January 1, 2027. Batch: 50% off | ai.google.dev |
| Gemini 3.5 Flash-Lite (Google) | $0.30 | $0.03 | $2.50 | Batch: 50% off | ai.google.dev |
| Grok 4.7 (xAI) | $2.00 | $0.50 | $6.00 | Prompts over 200K tokens: $4.00 in, $12.00 out. Batch: 20% off | docs.x.ai |
| DeepSeek Flash (DeepSeek) | $0.30 | $0.006 | $1.20 | No batch discount. Peak-hour price. Off-peak hours cost half. | api-docs.deepseek.com |
| DeepSeek V4 Pro (DeepSeek) | $1.32 | $0.044 | $3.96 | No batch discount. Peak-hour price. Off-peak hours cost half. | api-docs.deepseek.com |
How it’s calculated
Cost per request is input tokens times the input price plus output tokens times the output price, with prices per million tokens. The cached share of input is billed at the cached rate. Where a provider charges more for long prompts, the higher rate applies to the whole request once the prompt passes the threshold, as their pages state. Batch discounts apply to both input and output.
We check every listed price against the provider’s page each week. Writing to the prompt cache, web search, tools, images and audio are billed separately and are not included. Models count tokens differently, so the same text can be a different number of tokens on each.
Worked example
A support bot with 2,000 input and 500 output tokens per reply on Claude Sonnet 5.5 costs 2,000 × $2 + 500 × $10 per million, or $0.009 a reply. At 100,000 replies a month that is $900.