What will your LLM API bill actually be?
Pick a model, enter your token usage, and see the real cost in seconds — official prices only, verified 2026-10-27, no sign-up. Stop guessing before the invoice arrives.
Paste text to count tokens instead
Cost per call
—
Estimated monthly cost
—
| Model | Per call | Per month |
|---|
Need tokens only? Use the free AI token counter — same local counting, no cost fields.
Official price vs gateway price
The OpenAI, Anthropic and Google models tracked here are priced above at the official list rate. We have not found a gateway that sells them cheaper with a verifiable public price, so no gateway row is shown — we never estimate one.
Using open-weight models instead (DeepSeek, Qwen, Llama, gpt-oss)? See our Novita AI gateway prices, or go toNovita AI (affiliate).Disclosure.
Current official prices
USD per 1M tokens. Prices verified 2026-10-27 — see each model's page for its own verification date.
| Model | Provider | Input | Output | Cached input | Verified |
|---|---|---|---|---|---|
| GPT-6 Astra | OpenAI | $10 | $50 | $1 | 2026-10-25 |
| GPT-6.1 Sol | OpenAI | $2 | $10 | $0.1 | 2026-10-25 |
| Claude Sonnet 5.5 | Anthropic | $2 | $10 | $0.1 | 2026-10-25 |
| Claude Haiku 5.5 | Anthropic | $0.1 | $0.5 | $0.01 | 2026-10-25 |
| Gemini 4 Argon | $2 | $10 | $0.1 | 2026-10-24 | |
| GPT-6 Luna | OpenAI | $0.1 | $0.5 | $0.01 | 2026-10-25 |
| Claude Opus 5.5 | Anthropic | $4 | $20 | $0.2 | 2026-10-25 |
| Gemini 3.1 Pro | $2 | $12 | $0.2 | 2026-10-25 | |
| Gemini 3.8 Flash | $0.75 | $3.75 | $0.075 | 2026-10-25 | |
| Gemini 3.1 Flash-Lite | $0.25 | $1.5 | $0.025 | 2026-10-25 | |
| DeepSeek V4 Pro | DeepSeek | $1.32 | $3.96 | $0.044 | 2026-10-26 |
| DeepSeek V4.1 Flash | DeepSeek | $0.3 | $1.2 | $0.006 | 2026-10-26 |
| GPT-6 Sol | OpenAI | $2 | $10 | $0.2 | 2026-10-25 |
| GPT-5.6 Sol | OpenAI | $4 | $20 | $0.4 | 2026-10-25 |
| GPT-5.6 Terra | OpenAI | $2 | $12 | $0.2 | 2026-10-25 |
| GPT-5.6 Luna | OpenAI | $0.2 | $1.2 | $0.02 | 2026-10-25 |
| Claude Opus 5 | Anthropic | $5 | $25 | $0.5 | 2026-10-25 |
| Claude Sonnet 5 | Anthropic | $2 | $10 | $0.2 | 2026-10-25 |
| Gemini 3.6 Flash | $0.75 | $3.75 | $0.075 | 2026-10-26 | |
| Gemini 3.5 Flash-Lite | $0.3 | $2.5 | $0.03 | 2026-10-26 | |
| Grok 4.7 | xAI | $2 | $6 | $0.5 | 2026-10-26 |
| Grok 4.6 | xAI | $2 | $6 | $0.5 | 2026-10-26 |
| Grok 4.5 | xAI | $2 | $6 | $0.3 | 2026-10-26 |
| Grok 4.3 | xAI | $1.25 | $2.5 | $0.2 | 2026-10-26 |
| Kimi K3.1 Pro | Kimi | $2 | $10 | $0.2 | 2026-10-26 |
| Kimi K3.1 Pro Highspeed | Kimi | $4 | $20 | $0.2 | 2026-10-26 |
| Kimi K3 | Kimi | $3 | $15 | $0.3 | 2026-10-26 |
| GPT-5 Mini | OpenAI | $0.25 | $2 | $0.025 | 2026-10-26 |
| GPT-4o Mini | OpenAI | $0.15 | $0.6 | $0.075 | 2026-10-26 |
| GPT-4o | OpenAI | $2.5 | $10 | $1.25 | 2026-10-26 |
| GPT-4.1 | OpenAI | $2 | $8 | $0.5 | 2026-10-26 |
| GPT-4.1 Mini | OpenAI | $0.4 | $1.6 | $0.1 | 2026-10-26 |
| Claude Sonnet 4.5 | Anthropic | $3 | $15 | $0.3 | 2026-10-26 |
| Gemini 2.5 Flash | $0.3 | $2.5 | $0.03 | 2026-10-26 | |
| Gemini 2.5 Pro | $1.25 | $10 | $0.125 | 2026-10-26 | |
| Kimi K2.6 | Kimi | $0.95 | $4 | $0.16 | 2026-10-27 |
| Kimi K2.7 Code | Kimi | $0.95 | $4 | $0.19 | 2026-10-27 |
| Kimi K2.7 Code Highspeed | Kimi | $1.9 | $8 | $0.38 | 2026-10-27 |
| Mistral Large 4 | Mistral | $1.36 | $4.18 | $0.14 | 2026-10-27 |
| Mistral Medium 3.5 | Mistral | $1.5 | $7.5 | $0.15 | 2026-10-27 |
| Claude Fable 5.1 | Anthropic | $10 | $50 | $0.25 | 2026-10-27 |
| Claude Fable 5 | Anthropic | $10 | $50 | $1 | 2026-10-27 |
| Perplexity Sonar | Perplexity | $0.25 | $2.5 | $0.0625 | 2026-10-27 |
Popular model calculators
Browse providers or head-to-head comparisons.
Frequently asked questions
How accurate are these prices?
Every price comes from the provider’s official pricing page and is cross-checked against a second source. Each model shows the date its price was last verified, and the full dataset is re-checked weekly. Where a provider has not published a figure, we show “Not published” rather than inventing one.
What is the difference between official and gateway prices?
Official prices are what OpenAI, Anthropic, Google and other makers charge for direct API access. Gateways such as OpenRouter resell access at their own rates, which can differ. This calculator defaults to official prices and always labels the source — many calculators mix the two without telling you.
How does the cost calculator work?
Pick a model, enter input / output / cached token counts (or paste text to estimate tokens), and optionally set call volume. Cost = (input × input price + output × output price + cached × cached price) ÷ 1,000,000, then multiplied by your call count. All math runs in your browser; your text is never uploaded.
How are tokens counted when I paste text?
For OpenAI models we count with the real o200k tokenizer in your browser. Anthropic and Google publish no official browser tokenizer for current models, so those counts are labeled estimates. Your text is never uploaded anywhere.
How do you rank models, and do affiliates change the order?
Rankings and comparisons on this site are sorted by price only. Some links are affiliates (currently Novita AI), marked “(affiliate)” and disclosed on the About page. Commissions never change a listed price, a calculator result, or ranking order.