All calculators
Software factory and AI coding

Code modernization token estimator: repo size to tokens

Turn the size of a repository into the input and output tokens an agent needs to read and rewrite it, and the cost at your model price

Your numbers

Result

Estimated token cost

$70

Tokens in the codebase
2,500,000
Input tokens
20,000,000
Output tokens
3,000,000

LLM API prices per 1M tokens

Standard prices for current Claude, GPT and Gemini models in US dollars per million tokens. Pick a model above to fill the calculator with its prices.

Model Provider Input Cached input Output
Claude Fable 5.1 Anthropic $10.00 $0.25 $50.00
Claude Opus 5.5 Anthropic $4.00 $0.20 $20.00
Claude Sonnet 5.5 Anthropic $2.00 $0.10 $10.00
Claude Haiku 5.5 Prompts up to 100K tokens Anthropic $0.10 $0.01 $0.50
GPT-6 Astra Up to 272K input tokens OpenAI $10.00 $1.00 $50.00
GPT-6.1 Sol Up to 272K input tokens OpenAI $2.00 $0.10 $10.00
GPT-6.1 Terra Up to 272K input tokens OpenAI $1.00 $0.05 $5.00
GPT-6 Luna Up to 272K input tokens OpenAI $0.10 $0.01 $0.50
Gemini 4 Argon Introductory price until 30 Nov 2026, then $4 / $20 Google $2.00 $0.10 $10.00
Gemini 3.1 Pro (preview) Prompts up to 200K tokens Google $2.00 $0.20 $12.00
Gemini 3.8 Flash Introductory price until 31 Dec 2026, then $1.50 / $7.50 Google $0.75 $0.075 $3.75
Gemini 3.5 Flash-Lite Google $0.30 $0.03 $2.50

Prices checked on November 3, 2026 from the official pricing pages: Anthropic, OpenAI, Google. Batch, priority and regional pricing differ. Always confirm on the provider’s page before you budget.

How it is calculated

Code tokens = lines × tokens per line. Input tokens = code tokens × read passes (agents re-read files and context). Output tokens = code tokens × share rewritten × output overhead. Cost = input × input price + output × output price, per 1M tokens.

Keep in mind

The default prices are Claude Sonnet 5.5 list prices checked on 3 November 2026. Prices change, so confirm them on your provider’s pricing page.

Related service

Agent platform consultancy

Want these numbers worked out for your own process?

Let's talk