LLM API Cost Calculator
Work out what an LLM API workload costs from input tokens, output tokens and the number of requests, priced at current OpenAI, Anthropic Claude and Google Gemini list prices.
Last updated: October 2026
Total API cost
$90.00
Formula below · 3 sources (developers.openai.com, platform.claude.com, ai.google.dev) · Updated Oct 2026
Compare with similar
About this calculator
API providers bill input tokens (everything you send: system prompt, instructions, documents, chat history) and output tokens (everything the model writes, including any reasoning or "thinking" tokens) at separate prices per million tokens:
cost = requests × (input tokens × input price + output tokens × output price) ÷ 1,000,000
Output tokens cost several times as much as input tokens (5× for every Claude and GPT-6 model listed, about 8× for Gemini 3.5 Flash-Lite), so long answers can dominate the bill even when the prompt is much longer.
The model list carries each provider’s standard list price as of October 2026. Two price rules are built in: - OpenAI GPT-6 models: a request whose prompt is over 272,000 input tokens is billed at 2× the input price and 1.5× the output price for the whole request. - Claude Haiku 5.5: a request whose prompt is over 100,000 tokens pays $0.50 / $2.50 instead of $0.10 / $0.50.
Google has announced increases for Gemini 4 Argon (December 1, 2026) and Gemini 3.8 Flash (January 1, 2027), so both the current and the announced prices are listed.
Not included: prompt caching and Batch API discounts (see the prompt caching and batch calculators), US-only or regional processing surcharges (10% on some models), tool fees such as web search, and taxes.
| Model | Input | Output | Note |
|---|---|---|---|
| GPT-6 Astra | $10 | $50 | $20 / $75 over 272K-token prompts |
| GPT-6.1 Sol | $2 | $10 | $4 / $15 over 272K-token prompts |
| GPT-6 Luna | $0.10 | $0.50 | $0.20 / $0.75 over 272K-token prompts |
| Claude Fable 5.1 | $10 | $50 | |
| Claude Opus 5.5 | $4 | $20 | |
| Claude Sonnet 5.5 | $2 | $10 | |
| Claude Haiku 5.5 | $0.10 | $0.50 | $0.50 / $2.50 over 100K-token prompts |
| Gemini 4 Argon | $2 | $10 | to Nov 30, 2026 |
| Gemini 4 Argon | $4 | $20 | from Dec 1, 2026 |
| Gemini 3.8 Flash | $0.75 | $3.75 | to Dec 31, 2026 |
| Gemini 3.8 Flash | $1.50 | $7.50 | from Jan 1, 2027 |
| Gemini 3.5 Flash-Lite | $0.30 | $2.50 |
How to use
Pick a model, enter the average input and output tokens per request (the token counter converts words), and the number of requests. Example (the defaults): Claude Sonnet 5.5 at $2 / $10, 2,000 input and 500 output tokens, 10,000 requests. Input: 10,000 × 2,000 × $2 ÷ 1M = $40. Output: 10,000 × 500 × $10 ÷ 1M = $50. Total $90.00, or $0.009 per request. Example 2: the same workload on GPT-6 Luna ($0.10 / $0.50): 10,000 × (2,000 × 0.10 + 500 × 0.50) ÷ 1M = $4.50. Example 3: 300,000-token prompts on GPT-6.1 Sol cross the 272K line, so 100 requests with 1,000 output tokens cost 100 × (300,000 × $4 + 1,000 × $15) ÷ 1M = $121.50.
Frequently asked questions
How do I estimate tokens per request?
Add up everything sent on each call: system prompt, instructions, retrieved documents and the conversation so far, then the expected reply. For English, OpenAI’s rule of thumb is about 4 characters or ¾ of a word per token; current Claude models use about 1.8 tokens per word. The API response reports exact usage, so measure a sample of real requests once you can.
Why are output tokens more expensive?
The model generates output one token at a time, which takes far more compute per token than reading the prompt. At the prices listed here output costs 5× the input rate for every Claude and GPT-6 model and between 5× and about 8× for the Gemini models.
Do reasoning or thinking tokens count?
Yes. Reasoning models bill their internal reasoning as output tokens even when you do not see it. Google’s price list says “output price (including thinking tokens)”. Lower reasoning effort, or a non-reasoning setting where the model offers one, reduces the output count.
How current are the prices?
They are the standard list prices on the OpenAI, Anthropic and Google pricing pages as of October 2026 (checked 17 October 2026), before taxes, volume discounts or regional surcharges. Google has announced increases for Gemini 4 Argon (from December 1, 2026) and Gemini 3.8 Flash (from January 1, 2027); both prices are in the list. Providers change prices often, so confirm on the official pricing page before you commit a budget.
Related calculators
Sources & references
Think a result is wrong? Report a problem with this calculator.