Estimate tokens, compare API costs across 20+ current models with dated, sourced list prices, and check context-window fit. Free, no signup, runs entirely in your browser - nothing you type leaves this page.
Paste any text. Counts update as you type.
This is an estimate, not exact tokenization. It blends a characters/4 heuristic with a words-based ratio, which tracks byte-pair-encoding (BPE) tokenizers to within roughly 10 percent for typical English prose. Code, non-English text, and unusual formatting can deviate more. Each provider's tokenizer differs - use their official token-counting APIs for exact counts.
Published list prices per 1M tokens, as of 2026-08-27. Sources below.
List prices only. Excludes prompt-caching discounts, batch discounts, free tiers, and volume pricing - actual bills can be lower.
Same workload, every model. Selected model highlighted. Sorted by monthly cost.
| Model | Input $/1M | Output $/1M | Per request | Per day | Per month |
|---|
Uses your text from the estimator above and the selected model.
Context limits are as published by each provider; a real request also needs room for output tokens and system prompts.
Weio routes tasks to the most cost-efficient model that passes quality - connected to your actual tools. Stop guessing which model each job needs.
Try Weio freeEvery price on this page is a published list price, checked against the provider's official pricing page on 2026-08-27:
Notes: Gemini 3.7 Flash and 3.6 Flash prices are promotional through 2026-12-31 (rising to $1.50 / $7.50 on 2027-01-01). Gemini 2.5 Pro prices shown are for prompts up to 200K tokens ($2.50 / $15.00 above that). Llama and Qwen are open-weight models - prices shown are Together AI's hosted rates; other hosts differ. Prices change; if this page's date looks stale, check the source links.