Benchmarks / Cost Calculator

LLM Cost-per-Word Calculator

Paste your text, pick a model, see what it actually costs per word. Tokenizer efficiency (E) values come from the 50-model benchmark — not the marketing price sheet.

Your text (input prompt)

Model (sets tokenizer efficiency E)

Input price ($ / million tokens)

Calls per month

—

Words in your text

—

Est. input tokens (E × words)

—

Cost per 1,000 words

—

Est. monthly cost

How this works: word count uses Unicode text segmentation (the same standard the benchmark uses). Estimated tokens = words × E, where E is the model's measured average tokens-per-word on a 60/40 prose/code blend. Real token counts vary by content (code tokenizes differently than prose — see the benchmark's code-E vs prose-E chart). This is a planning estimate, not an invoice.

Same text across all 50 models

Ranked by cost per 1,000 words at the price you entered above. Green = cheapest 25%, red = priciest 25%.

E values are measured, not claimed. Full methodology, raw CSV, and the harness that produced these numbers: github.com/ClockLobsterLabs/LLM-Cost-Comparison. Prices are user-entered and age fast — verify against the provider's current pricing page before committing.