AI Token Cost Calculator
AI Cost Calculator estimates what an LLM API call costs from token counts and per-million pricing for GPT, Claude, Gemini and more, with no key and no account.
Cost per 1M tokens: GPT-6 Astra costs $10.00 per 1M input tokens and $50.00 per 1M output tokens. Across all 65 models listed, 1M input tokens cost $0.05 to $30.00 and 1M output tokens $0.40 to $180.00. Official list prices, last checked .
gpt-6-astra | Input: $10.00 per 1M tokens | Output: $50.00 per 1M tokens- Input cost
- $0.010000
- Output cost
- $0.025000
- Total cost
- $0.035000
About LLM and AI model pricing
Providers charge per token, quoted per 1 million tokens, with separate rates for input (your prompt) and output (the response). Output usually costs several times more than input, so a chatty response can cost more than a long prompt.
A token is roughly 4 characters or about 3/4 of a word in English, but each provider's tokenizer differs, so the same text can be a different number of tokens on different models.
LLM API pricing comparison
| Model | Status | Input / 1M tokens | Output / 1M tokens |
|---|---|---|---|
| GPT-6 Astragpt-6-astra | Current | $10.00 | $50.00 |
| GPT-6.1 Solgpt-6.1-sol | Current | $2.00 | $10.00 |
| GPT-6 Lunagpt-6-luna | Current | $0.10 | $0.50 |
| GPT-6 Solgpt-6-sol | Available | $2.00 | $10.00 |
| GPT-5.6 Solgpt-5.6-sol | Available | $4.00 | $20.00 |
| GPT-5.6 Terragpt-5.6-terra | Available | $2.00 | $12.00 |
| GPT-5.6 Lunagpt-5.6-luna | Available | $0.20 | $1.20 |
| GPT-5.5gpt-5.5 | Available | $5.00 | $30.00 |
| GPT-5.5 Progpt-5.5-pro | Available | $30.00 | $180.00 |
| GPT-5.4gpt-5.4 | Available | $2.50 | $15.00 |
| GPT-5.4 Minigpt-5.4-mini | Available | $0.75 | $4.50 |
| GPT-5.4 nanogpt-5.4-nano | Deprecated | $0.20 | $1.25 |
| GPT-5.4 Progpt-5.4-pro | Available | $30.00 | $180.00 |
| GPT-5.2gpt-5.2 | Available | $1.75 | $14.00 |
| GPT-5.2 Progpt-5.2-pro | Available | $21.00 | $168.00 |
| GPT-5.1gpt-5.1 | Deprecated | $1.25 | $10.00 |
| GPT-5gpt-5 | Deprecated | $1.25 | $10.00 |
| GPT-5 Minigpt-5-mini | Deprecated | $0.25 | $2.00 |
| GPT-5 nanogpt-5-nano | Deprecated | $0.05 | $0.40 |
| GPT-5 Progpt-5-pro | Deprecated | $15.00 | $120.00 |
| GPT-4.1gpt-4.1 | Available | $2.00 | $8.00 |
| GPT-4.1 Minigpt-4.1-mini | Available | $0.40 | $1.60 |
| GPT-4.1 nanogpt-4.1-nano | Deprecated | $0.10 | $0.40 |
| GPT-4ogpt-4o | Available | $2.50 | $10.00 |
| GPT-4o Minigpt-4o-mini | Available | $0.15 | $0.60 |
| o3o3 | Deprecated | $2.00 | $8.00 |
| o3-proo3-pro | Deprecated | $20.00 | $80.00 |
| o4-minio4-mini | Deprecated | $1.10 | $4.40 |
| o3-minio3-mini | Deprecated | $1.10 | $4.40 |
| o1o1 | Deprecated | $15.00 | $60.00 |
| Claude Fable 5.1claude-fable-5-1 | Current | $10.00 | $50.00 |
| Claude Opus 5.5claude-opus-5-5 | Current | $4.00 | $20.00 |
| Claude Sonnet 5.5claude-sonnet-5-5 | Current | $2.00 | $10.00 |
| Claude Haiku 5.5claude-haiku-5-5 | Current | $0.10 | $0.50 |
| Claude Fable 5claude-fable-5 | Legacy | $10.00 | $50.00 |
| Claude Opus 5claude-opus-5 | Legacy | $5.00 | $25.00 |
| Claude Opus 4.8claude-opus-4-8 | Legacy | $5.00 | $25.00 |
| Claude Opus 4.7claude-opus-4-7 | Legacy | $5.00 | $25.00 |
| Claude Opus 4.6claude-opus-4-6 | Legacy | $5.00 | $25.00 |
| Claude Opus 4.5claude-opus-4-5-20251101 | Legacy | $5.00 | $25.00 |
| Claude Sonnet 5claude-sonnet-5 | Legacy | $2.00 | $10.00 |
| Claude Sonnet 4.6claude-sonnet-4-6 | Legacy | $3.00 | $15.00 |
| Claude Sonnet 4.5claude-sonnet-4-5-20250929 | Deprecated | $3.00 | $15.00 |
| Claude Haiku 4.5claude-haiku-4-5-20251001 | Legacy | $1.00 | $5.00 |
| Gemini 3.8 Flashgemini-3.8-flash | Current | $0.75 | $3.75 |
| Gemini 3.5 Flash-Litegemini-3.5-flash-lite | Current | $0.30 | $2.50 |
| Gemini 3.7 Flashgemini-3.7-flash | Available | $0.75 | $3.75 |
| Gemini 3.6 Flashgemini-3.6-flash | Available | $0.75 | $3.75 |
| Gemini 3.5 Flashgemini-3.5-flash | Available | $1.50 | $9.00 |
| Gemini 3.1 Flash-Litegemini-3.1-flash-lite | Available | $0.25 | $1.50 |
| Gemini 3.1 Pro (preview)gemini-3.1-pro-preview | Preview | $2.00 | $12.00 |
| Gemini 3 Flash (preview)gemini-3-flash-preview | Preview | $0.50 | $3.00 |
| Gemini 2.5 Progemini-2.5-pro | Legacy | $1.25 | $10.00 |
| Gemini 2.5 Flashgemini-2.5-flash | Legacy | $0.30 | $2.50 |
| Gemini 2.5 Flash-Litegemini-2.5-flash-lite | Legacy | $0.10 | $0.40 |
| Grok 4.7grok-4.7 | Current | $2.00 | $6.00 |
| Grok 4.6grok-4.6 | Available | $2.00 | $6.00 |
| Grok 4.5grok-4.5 | Available | $2.00 | $6.00 |
| Grok 4.3grok-4.3 | Available | $1.25 | $2.50 |
| Grok 4.20 (reasoning)grok-4.20-0309-reasoning | Available | $1.25 | $2.50 |
| Grok 4.20 (non-reasoning)grok-4.20-0309-non-reasoning | Available | $1.25 | $2.50 |
| Grok 4.20 Multi-Agentgrok-4.20-multi-agent-0309 | Available | $1.25 | $2.50 |
| Grok Build 0.1grok-build-0.1 | Available | $1.00 | $2.00 |
| DeepSeek V4.1 Flashdeepseek-flash | Current | $0.30 | $1.20 |
| DeepSeek V4 Prodeepseek-v4-pro | Current | $1.32 | $3.96 |
Standard prices per 1 million tokens from each provider's pricing page, last checked October 8, 2026. Batch, cached-input and long-prompt rates are not shown here; see the model ID lookup for the notes on each model.
At a glance
- Accepts
- Token counts and a model you pick
- Outputs
- Estimated cost per call and per million tokens
- Where it runs
- Entirely in your browser, on your device
- What leaves your device
- Nothing - what you type stays in this tab
- Limits
- An estimate from listed prices, not a billing figure
- Account
- Not required - there is no sign-up
- Price
- Free, with no watermark and no usage cap
- Made by
- Visrut
- Last updated
- 8 October 2026
Pick a model from OpenAI, Anthropic, Google, xAI or DeepSeek, enter how many input and output tokens you expect, and see the cost at once. The math, (tokens / 1,000,000) x price per million, runs in your browser against the same official price list as the AI model ID lookup; no API key, account or usage data leaves your machine.
How to use it
- Choose an AI model from the dropdown.
- Enter expected input and output token counts, or click a preset from 500 tokens up to 1M tokens.
- Read the input, output and total cost. They update as you type, and switch to the long-prompt rate when your input passes the model's threshold.
- Click "Share this calculation" to copy a URL that reproduces the same model and token counts.
Frequently asked questions
How much do 1M AI tokens cost?
It depends on the model and on whether the tokens are input or output. Across the 65 models here, 1M input tokens cost from $0.05 (GPT-5 nano) to $30.00 (GPT-5.4 Pro), and 1M output tokens from $0.40 to $180.00. Output typically costs 3 to 8 times the input price. Prices checked October 8, 2026.
Does this send my token counts or usage anywhere?
No. The whole calculation is (tokens / 1,000,000) x price per million, run in your browser. Nothing you enter is transmitted anywhere.
Where do the prices come from?
From each provider's official pricing page (OpenAI, Anthropic, Google, xAI and DeepSeek), last checked October 8, 2026. They are the same numbers shown on the AI model ID lookup, from one shared list, so the two pages always agree. They are standard rates: batch discounts and cached-input prices are not applied.
Why did the price per token go up when I entered a big prompt?
Several providers charge more once a single prompt passes a size threshold, and then bill the whole request at the higher rate: over 272K input tokens on recent OpenAI models, over 200K on Gemini 3.1 Pro preview and Gemini 2.5 Pro, 200K or more on Grok, and over 100K on Claude Haiku 5.5. The calculator switches to that rate automatically and says so.
What do input and output tokens mean?
Input tokens are what you send the model (the prompt, including any files or chat history); output tokens are what it generates back, including reasoning or thinking tokens on models that bill them as output. Providers price the two differently, which is why the calculator asks for both.
What does the Share this calculation button do?
It copies the current page URL, with your model ID and token counts as query parameters. Opening that link reproduces the same numbers. It is just a URL, not a record saved on any server.