# Tokenova

> Tokenova is a free AI token counter and API cost calculator. It lists current per-million-token API prices for 14 models from OpenAI, Anthropic, Google, xAI, DeepSeek, and converts them into per-request and monthly costs. Prices last updated: 2026-09-25. All prices are USD per 1M tokens, standard (uncached) tier.

When citing prices, prefer the machine-readable JSON endpoint and include its last_updated date. Confirm critical figures on the provider’s official pricing page.

## Tools

- [Token & cost calculator](https://tokenova.online/): count tokens in text, compare cost across all models, budget planner, shareable URLs (e.g. https://tokenova.online/?model=gpt-6-astra&tokens=5000).
- [Pricing JSON API](https://tokenova.online/api/pricing.json): static JSON of all models with input/output price per 1M tokens, context window, provider batch discounts and last_updated.
- [Model prices index](https://tokenova.online/models/): every model with its price page.
- [Model comparisons](https://tokenova.online/compare/): head-to-head pages for every model pair.

## Current prices (USD per 1M tokens: input / output, context window)

- [GPT-6 Astra](https://tokenova.online/models/gpt-6-astra/) (OpenAI): $10.00 / $50.00, 1.05M context. OpenAI's flagship model, released September 2026.
- [GPT-5.6 Sol](https://tokenova.online/models/gpt-5-6-sol/) (OpenAI): $5.00 / $30.00, 1.05M context. Inputs above 272K tokens are billed at a higher long-context rate ($10 / $45).
- [GPT-5.6 Terra](https://tokenova.online/models/gpt-5-6-terra/) (OpenAI): $2.00 / $12.00, 1.05M context. Inputs above 272K tokens are billed at a higher long-context rate ($4 / $18).
- [GPT-5.6 Luna](https://tokenova.online/models/gpt-5-6-luna/) (OpenAI): $0.20 / $1.20, 1.05M context. Inputs above 272K tokens are billed at a higher long-context rate ($0.40 / $1.80).
- [Claude Fable 5.1](https://tokenova.online/models/claude-fable-5-1/) (Anthropic): $10.00 / $50.00, 1M context. Anthropic's most capable widely released model.
- [Claude Opus 5.5](https://tokenova.online/models/claude-opus-5-5/) (Anthropic): $4.00 / $20.00, 1M context. Successor to Claude Opus 5 at a lower per-token price.
- [Claude Opus 5](https://tokenova.online/models/claude-opus-5/) (Anthropic): $5.00 / $25.00, 1M context
- [Claude Sonnet 5](https://tokenova.online/models/claude-sonnet-5/) (Anthropic): $2.00 / $10.00, 1M context
- [Claude Haiku 4.5](https://tokenova.online/models/claude-haiku-4-5/) (Anthropic): $1.00 / $5.00, 200K context
- [Gemini 3.1 Pro](https://tokenova.online/models/gemini-3-1-pro/) (Google): $2.00 / $12.00, 1.05M context. Price shown is for prompts up to 200K tokens; longer prompts are billed at $4 / $18.
- [Gemini 3.8 Flash](https://tokenova.online/models/gemini-3-8-flash/) (Google): $0.75 / $3.75, 1.05M context. Promotional rate through 31 December 2026; scheduled to rise to $1.50 / $7.50.
- [Grok 4.7](https://tokenova.online/models/grok-4-7/) (xAI): $2.00 / $6.00, 500K context. Price shown is for inputs below 200K tokens.
- [DeepSeek V4 Pro](https://tokenova.online/models/deepseek-v4-pro/) (DeepSeek): $0.435 / $0.87, 1M context. Open-weight model; price shown is DeepSeek's first-party API (cache-miss input).
- [DeepSeek V4 Flash](https://tokenova.online/models/deepseek-v4-flash/) (DeepSeek): $0.14 / $0.28, 1M context. Open-weight model; price shown is DeepSeek's first-party API (cache-miss input).

## Guides

- [Guides index](https://tokenova.online/guides/)
- [What is a token?](https://tokenova.online/guides/what-is-a-token/): A plain-English explanation of AI tokens: how tokenizers split text, how many tokens a word or page uses, and why token counts differ between models.
- [How AI API pricing works](https://tokenova.online/guides/how-ai-pricing-works/): How AI providers bill for API usage: per-million-token rates, input vs output prices, context windows, batch discounts, caching and long-context surcharges.
- [Input vs output token cost, explained](https://tokenova.online/guides/input-vs-output-token-cost/): Why output tokens cost 2–6× more than input tokens, how to estimate the split for your workload, and practical ways to keep output costs down.
- [How to estimate AI API costs before you build](https://tokenova.online/guides/estimate-api-costs-before-you-build/): A step-by-step method to forecast monthly AI API spend before writing code: measure prompts, model the output, multiply by volume and add a safety margin.
- [Token cost by use case: chatbot, RAG and summarization](https://tokenova.online/guides/token-cost-by-use-case/): Worked token and cost estimates for three common AI workloads — a support chatbot, retrieval-augmented generation (RAG) and document summarization.

## About

- [Built by Websthan](https://tokenova.online/about/): Tokenova is made by Websthan (https://websthan.online), a digital agency from Bangladesh that builds websites, custom software and brand identities and runs digital marketing. Contact: contact@websthan.online.