AI token cost calculator
What your input and output tokens cost on every model we track, cheapest first, at today's list prices.
Result
| Model | Your cost | Input / 1M | Output / 1M | Cached / 1M |
|---|---|---|---|---|
| Claude Haiku 5.5Anthropic | $0.57 | $0.10 | $0.50 | $0.01 |
| GPT-6 LunaOpenAI | $0.57 | $0.10 | $0.50 | $0.01 |
| GLM-5.3 FlashZ.ai | $0.82 | $0.15 | $0.50 | $0.03 |
| gpt-oss-120bOpenAI | $0.84 | $0.15 | $0.60 | $0.075 |
| MiniMax M3.1 FlashMiniMax | $0.84 | $0.15 | $0.60 | $0.015 |
| DeepSeek V4.1 FlashDeepSeek | $1.68 | $0.30 | $1.20 | $0.006 |
| MiniMax M3MiniMax | $1.68 | $0.30 | $1.20 | $0.06 |
| Gemini 3.5 Flash-LiteGoogle | $1.88 | $0.30 | $2.50 | $0.03 |
| MiMo-V2.6-ProXiaomi | $2.31 | $0.435 | $0.87 | $0.0036 |
| Qwen3.8 27BAlibaba (Qwen) | $2.51 | $0.425 | $2.55 | $0.085 |
| DeepSeek V4 ProDeepSeek | $3.60 | $0.66 | $1.98 | $0.022 |
| Gemini 3.6 FlashGoogle | $4.31 | $0.75 | $3.75 | $0.075 |
| Gemini 3.8 FlashGoogle | $4.31 | $0.75 | $3.75 | $0.075 |
| Kimi K2.6Moonshot AI | $5.35 | $0.95 | $4 | $0.16 |
| Claude Haiku 4.5Anthropic | $5.75 | $1 | $5 | $0.10 |
| Grok 4.3xAI | $6.63 | $1.25 | $2.50 | $0.20 |
| Muse Spark OpenMeta | $6.89 | $1.25 | $4.25 | $0.15 |
| Mistral Large 4Mistral | $7.43 | $1.36 | $4.18 | $0.14 |
| GLM-5.3Z.ai | $7.66 | $1.40 | $4.40 | $0.26 |
| Mistral Medium 3.5Mistral | $8.63 | $1.50 | $7.50 | n/a |
| Grok 4.6xAI | $10.90 | $2 | $6 | $0.50 |
| Grok 4.7xAI | $10.90 | $2 | $6 | $0.50 |
| Qwen3.8 MaxAlibaba (Qwen) | $10.90 | $2 | $6 | $0.25 |
| Claude Sonnet 5Anthropic | $11.50 | $2 | $10 | $0.20 |
| Claude Sonnet 5.5Anthropic | $11.50 | $2 | $10 | $0.10 |
| Gemini 4 ArgonGoogle | $11.50 | $2 | $10 | $0.10 |
| GPT-6 SolOpenAI | $11.50 | $2 | $10 | $0.20 |
| GPT-6.1 SolOpenAI | $11.50 | $2 | $10 | $0.10 |
| Gemini 3.1 Pro PreviewGoogle | $11.80 | $2 | $12 | $0.20 |
| Claude Sonnet 4.6Anthropic | $17.25 | $3 | $15 | $0.30 |
| Kimi K3Moonshot AI | $17.25 | $3 | $15 | $0.30 |
| Claude Opus 5.5Anthropic | $23 | $4 | $20 | $0.20 |
| GPT-5.6 SolOpenAI | $23 | $4 | $20 | $0.40 |
| Muse WonderMeta | $23 | $4 | $20 | $0.20 |
| Claude Opus 4.6Anthropic | $28.75 | $5 | $25 | $0.50 |
| Claude Opus 4.7Anthropic | $28.75 | $5 | $25 | $0.50 |
| Claude Opus 4.8Anthropic | $28.75 | $5 | $25 | $0.50 |
| Claude Opus 5Anthropic | $28.75 | $5 | $25 | $0.50 |
| GPT-5.5OpenAI | $29.50 | $5 | $30 | $0.50 |
| Claude Fable 5Anthropic | $57.50 | $10 | $50 | $1 |
| Claude Fable 5.1Anthropic | $57.50 | $10 | $50 | $0.25 |
| GPT-6 AstraOpenAI | $57.50 | $10 | $50 | $1 |
How we work it out
Cost = input tokens × input price + output tokens × output price, with the cached share of input at the cache read price. Prices are per 1M tokens. Each model's prices, with their sources, are on its comparison pages.
Questions
Where do the prices come from?
From each maker's own pricing page, or for open models from the price the maker's own endpoint lists on OpenRouter. Every price links to the page we read it from and the day we read it.
Why does the size of one call matter?
Some models charge more for long prompts: Claude Haiku 5.5 above 100K tokens, Gemini 3.1 Pro above 200K and GPT-6 Luna above 272K. The calculator applies the higher price when one call's prompt is over the line.
Does it include tax or batch discounts?
No. These are standard list prices in US dollars before any tax. Batch APIs are usually half price; we leave them out because agents don't use them.
More tools
- Compare AI modelsEvery pair, side by side
- Context window calculatorWill my code fit in the context window?
- Prompt caching calculatorHow much prompt caching saves you
- Claude plan vs API calculatorSubscription or pay-as-you-go?
- Free limit reset trackerIs a free Claude or Codex reset live right now?
- Claude usage limit trackerDid the 5-hour or weekly limit go up or down?
- Claude Code usage dashboardYour tokens, models, cost and busiest hours, from one command
Numbers from the makers' own pages, checked by hand, last read . List prices in US dollars before tax. Spotted a mistake? office@shuriken.sh.
Made by Shuriken, the harness for coding agents. Its hot swapping moves your chats to another of your accounts or API keys when one nears its limit. Use only accounts you own and are entitled to use.
Try Shuriken on your own computerYou're signed in
Add your Claude, Codex and Gemini accounts once. Shuriken switches between them so your agents keep working.Your accounts, projects and chats are in the app.
Coming soon