Are these prices in US dollars?

Yes. Every provider in our data publishes its API and GPU prices in US dollars, and this page shows them exactly as published, with no conversion.

Why might my invoice differ from this estimate?

Your real token counts, cache hit rate and retry rate will differ from any estimate, and providers bill some extras this calculator leaves out, such as cache writes, tool calls and data transfer. Prices also change; check the date at the top of the page and the provider’s own pricing page before you commit a budget.

Is it cheaper to use an API or rent GPUs?

At low or uneven traffic, an API is almost always cheaper because you pay only for tokens you use. Renting GPUs starts to win only at high, steady volume, with an open-weight model that meets your quality bar. Our guide on API vs self-hosted GPUs walks through the break-even math with current prices.

LLM API and GPU cost calculator

Estimate your monthly spend on large language model APIs and cloud GPU rentals from your own token volumes or GPU hours, and compare every model side by side, in US dollars.

Prices last verified between November 10, 2026 and November 11, 2026 against each provider’s published price list.See every source

Estimate your AI costs

All amounts in US dollars, from each provider’s published price list.

Model

Your traffic

Requests per day

API calls, not users. An agent or chain may make several calls per user action.

Average input tokens per request

tokens

System prompt, chat history, retrieved documents and tool definitions together.

Average output tokens per request

tokens

Include reasoning tokens if the model bills them as output.

Discounts

Input read from the prompt cache

Share of input tokens billed at the cached input price.

Requests sent through a batch API

Work that can wait for an asynchronous batch job instead of an instant reply.

The starting figures are examples. Replace them with your own.

GPT-6 Astra estimate

Per month: $7,604; Average month (365 ÷ 12 days)
Per year: $91,250
Per 1,000 requests: $25.00
Per request: $0.025

Where the money goes (prompts up to 272k input tokens)
Price	Tokens per day	Cost per day	Share
Input	10,000,000	$100.00	40%
Output	3,000,000	$150.00	60%

Sources: developers.openai.com (checked November 10, 2026)

Compare every model for this workload

Same requests and token counts on each model, cheapest first. Select a column heading to sort, or a model name to see its breakdown above. List prices are the standard rates for the price tier your prompt length falls in.

Model costs for your workload, sorted by per month (ascending)
							vs cheapest
Prompts up to 100,000 input tokens	Anthropic	$0.10	$0.01	$0.50	$0.25	$76.04	Cheapest
Prompts up to 272K input tokens	OpenAI	$0.10	$0.01	$0.50	$0.25	$76.04	Cheapest
	Mistral AI	$0.15	n/a	$0.60	$0.33	$100.38	1.3×
	Google (Gemini API)	$0.30	$0.03	$2.50	$1.05	$319.38	4.2×
	Mistral AI	$0.70	$0.07	$2.10	$1.33	$404.54	5.3×
	Google (Gemini API)	$0.75	$0.075	$3.75	$1.88	$570.31	7.5×
Prompts up to 272K input tokens	OpenAI	$1.00	$0.05	$5.00	$2.50	$760.42	10.0×
	Anthropic	$2.00	$0.10	$10.00	$5.00	$1,521	20.0×
	Google (Gemini API)	$2.00	$0.10	$10.00	$5.00	$1,521	20.0×
Prompts up to 272K input tokens	OpenAI	$2.00	$0.10	$10.00	$5.00	$1,521	20.0×
	Anthropic	$4.00	$0.20	$20.00	$10.00	$3,042	40.0×
	Anthropic	$8.00	$0.20	$40.00	$20.00	$6,083	80.0×
Prompts up to 272K input tokens	OpenAI	$10.00	$1.00	$50.00	$25.00	$7,604	100.0×

Hardware

GPU

Provider

Where a provider sells several instance sizes, the cheapest one that fits your GPU count is used.

Pricing

A reservation is paid for every hour of the term, used or not.

Number of GPUs

Rounded up to whole instances where a provider only rents fixed sizes.

Running time and output

Hours switched on per day

hours

Busy share of those hours

How much of the switched-on time the GPUs spend generating at full throughput. Real services rarely reach 100%.

Throughput per GPU (optional)

tokens/s

No verified benchmark for this GPU yet. Enter a figure you have measured to see a cost per 1M tokens.

NVIDIA H100 80GB (SXM) estimate

Per month: $40,179; 730 billed hours
Per year: $482,150
Per hour: $55.04; 8 GPUs billed
Per 1M tokens: n/a; Needs a throughput figure

Priced on AWS EC2 p5.48xlarge (US East (N. Virginia), Linux), 1 × 8 GPUs.

Sources: aws.amazon.com (checked November 10, 2026); b0.p.awsstatic.com (checked November 10, 2026)

Every provider for the NVIDIA H100 80GB (SXM)

Monthly cost for 8 GPUs at 24 hours a day, cheapest on demand price first. Select a provider to see its breakdown.

On demand	Spot	1-year reserved
$15,710	No self-serve price published for this option.	No self-serve price published for this option.
$23,302	No self-serve price published for this option.	No self-serve price published for this option.
$26,806	No self-serve price published for this option.	No self-serve price published for this option.
$40,179	Price not verified yet.	Price not verified yet.
$71,774	$13,264	$45,935
Price not verified yet.	Price not verified yet.	Price not verified yet.

Guides

How to estimate LLM API costs before launch
A step-by-step method for pricing an LLM feature before it ships: define the request, measure tokens, model traffic, check price tiers and turn one guess into a low, expected and high range.
Prompt caching and batch APIs: what they actually save
Why a large cache discount on the price list turns into a smaller saving on the bill, when batch processing fits, and how to work out the real saving for your own traffic.
API vs self-hosted GPUs: the break-even math
When renting GPUs and running your own model costs less than paying per token: the break-even formula, the capacity ceiling, why idle time matters most, and when a reservation pays off.

All guides

Worked costs

By model
Each model at five standard workloads.
By GPU
Monthly rental cost across providers.
Comparisons
Models in the same tier, head to head.

Frequently asked questions

Are these prices in US dollars?: Yes. Every provider in our data publishes its API and GPU prices in US dollars, and this page shows them exactly as published, with no conversion.
Why might my invoice differ from this estimate?: Your real token counts, cache hit rate and retry rate will differ from any estimate, and providers bill some extras this calculator leaves out, such as cache writes, tool calls and data transfer. Prices also change; check the date at the top of the page and the provider’s own pricing page before you commit a budget.
Is it cheaper to use an API or rent GPUs?: At low or uneven traffic, an API is almost always cheaper because you pay only for tokens you use. Renting GPUs starts to win only at high, steady volume, with an open-weight model that meets your quality bar. Our guide on API vs self-hosted GPUs walks through the break-even math with current prices.