Home › Tools › Tokens per word: Spanish
Tokens per word in Spanish
Spanish averages about 1.7 tokens per word on current BPE tokenizers, which is 28% more tokens than the same count in English. Romance languages split into more sub-word pieces than English, so the same sentence costs roughly 28% more tokens.
Spanish to tokens table
| Words | Tokens |
|---|---|
| 100 | 170 |
| 250 | 425 |
| 500 | 850 |
| 1,000 | 1,700 |
| 2,500 | 4,250 |
| 5,000 | 8,500 |
| 10,000 | 17,000 |
| 25,000 | 42,500 |
| 50,000 | 85,000 |
| 100,000 | 170,000 |
Language comparison
| Language | Tokens per word | Tokens for 1,000 | Cost multiplier vs English |
|---|---|---|---|
| English | 1.33 | 1,330 | 1.00× |
| Spanish | 1.7 | 1,700 | 1.28× |
| French | 1.8 | 1,800 | 1.35× |
| German | 1.9 | 1,900 | 1.43× |
| Portuguese | 1.7 | 1,700 | 1.28× |
| Russian | 2.4 | 2,400 | 1.80× |
| Arabic | 2.9 | 2,900 | 2.18× |
| Hindi | 3.2 | 3,200 | 2.41× |
| Chinese | 1.5 | 1,500 | 1.13× |
| Japanese | 1.8 | 1,800 | 1.35× |
Ratios are per word except Chinese and Japanese, which are per character. Comparing them directly is approximate — the honest way to budget is to tokenize a sample of your own corpus.
Cost of 1,000 words in Spanish
| # | Model | Provider | Input | Output | Blended |
|---|---|---|---|---|---|
| 1 | GPT-6 Luna | OpenAI | $0.00017 | $0.00085 | $0.00038 |
| 2 | GPT-5.6 Luna | OpenAI | $0.00034 | $0.00204 | $0.00085 |
| 3 | GPT-5.4 nano | OpenAI | $0.00034 | $0.00212 | $0.00087 |
| 4 | Gemini 3.5 Flash-Lite | Google Gemini | $0.00051 | $0.00425 | $0.00157 |
| 5 | DeepSeek V4.1 Flash | DeepSeek | $0.00051 | $0.00204 | $0.00102 |
| 6 | GPT-5.4 mini | OpenAI | $0.00127 | $0.00765 | $0.00319 |
| 7 | Gemini 3.8 Flash | Google Gemini | $0.00127 | $0.00637 | $0.00287 |
| 8 | Gemini 3.7 Flash | Google Gemini | $0.00127 | $0.00637 | $0.00287 |
| 9 | Claude Haiku 4.5 | Anthropic | $0.0017 | $0.0085 | $0.00382 |
| 10 | Gemini 2.5 Pro | Google Gemini | $0.00212 | $0.017 | $0.00637 |
| 11 | Grok 4.3 | xAI Grok | $0.00212 | $0.00425 | $0.00319 |
| 12 | DeepSeek V4 Pro | DeepSeek | $0.00224 | $0.00673 | $0.00393 |
| 13 | Gemini 3.5 Flash | Google Gemini | $0.00255 | $0.0153 | $0.00637 |
| 14 | GPT-5.2 | OpenAI | $0.00297 | $0.0238 | $0.00892 |
| 15 | GPT-6.1 Sol | OpenAI | $0.0034 | $0.017 | $0.00765 |
| 16 | GPT-6 Sol | OpenAI | $0.0034 | $0.017 | $0.00765 |
| 17 | GPT-5.6 Terra | OpenAI | $0.0034 | $0.0204 | $0.0085 |
| 18 | GPT-4.1 | OpenAI | $0.0034 | $0.0136 | $0.0068 |
| 19 | Claude Sonnet 5 | Anthropic | $0.0034 | $0.017 | $0.00765 |
| 20 | Gemini 3.1 Pro (Preview) | Google Gemini | $0.0034 | $0.0204 | $0.0085 |
| 21 | Grok 4.6 | xAI Grok | $0.0034 | $0.0102 | $0.00595 |
| 22 | GPT-5.4 | OpenAI | $0.00425 | $0.0255 | $0.0106 |
| 23 | Claude Sonnet 4.6 | Anthropic | $0.0051 | $0.0255 | $0.0115 |
| 24 | GPT-5.6 Sol | OpenAI | $0.0068 | $0.034 | $0.0153 |
| 25 | GPT-5.5 | OpenAI | $0.0085 | $0.051 | $0.0212 |
| 26 | Claude Opus 5 | Anthropic | $0.0085 | $0.0425 | $0.0191 |
| 27 | GPT-6 Astra | OpenAI | $0.017 | $0.085 | $0.0382 |
| 28 | Claude Fable 5.1 | Anthropic | $0.017 | $0.085 | $0.0382 |
Input = 1,700 prompt tokens. Output = 1,700 generated tokens. Blended = input + 25% of output, the same weighting used across this site.
What this means for a multilingual product
If you serve Spanish and English from the same budget, Spanish consumes 1.3× the tokens for the same number of words. Either the price per token absorbs that, or the Spanish product quietly runs at a lower margin. Measuring the ratio on your own text is the only reliable way to price it.
FAQ
How many tokens is 1,000 words in Spanish?
About 1,700 tokens, using 1.7 tokens per word. That is 28% more tokens than the same count in English (1,330).
Why is Spanish more expensive per word?
Romance languages split into more sub-word pieces than English, so the same sentence costs roughly 28% more tokens. Providers bill per token, not per word, so a language that fragments into more tokens costs more for the same meaning.
How do I get an exact count for Spanish?
Run a sample of your own text through the tokenizer of the model you will be billed on. Public ratios are averages; product names, numbers and formatting in your own data can move the result by 20% or more.
Keep going
- Words to tokens (English baseline)
- Tokens to dollars
- Tokens per word: English
- Tokens per word: French
- Tokens per word: German
- All 28 models ranked by price
- Run your own volumes in the calculator