BroadRouter

Gemma 3 token counter

Updated

In short: This counter runs Google's official Gemma 3 tokenizer.json in your browser, pinned to the same commit as our tokenizer dataset. It has 262,145 token IDs and is a 33 MB download. English prose takes about 130 tokens per 100 words, Modern Standard Arabic about 203.

Free, no sign-up, runs in your browser: nothing you type is sent anywhere.

Try:

… tokens

Tokens and cost of this text per model
ModelTokensChars per tokenAs inputAs output
Gemma 3 4B
Gemma 3 tokenizer (Gemma 3 27B IT)
Price: Gemma 3 4B at Amazon Bedrock, cheapest listed input price (not BroadNet)
…………
DeepSeek V4.1 Flash
DeepSeek V4.1 Flash tokenizer
…………
GLM 5.3
GLM 5.3 tokenizer
…………
Qwen3.8 27B
Qwen3.8 27B tokenizer
…………
Gemma 4 VL 26B
Gemma 4 tokenizer (Google base model)
…………
GPT (o200k)
OpenAI o200k_base
…………

Cost = tokens × the model's published price per 1M input or output tokens. GPT is shown for reference only (no BroadNet price). Gemma 3 models are not sold by BroadNet: its row uses the cheapest standard input price we hold for a model with that tokenizer, read from the provider's own pricing page. Sources and check dates are in the real cost by language dataset. The chat-format option adds template tokens only for the DeepSeek, GLM, Qwen, Gemma and GPT rows; Gemma 3 row counts your text alone. Measured ratios for every tokenizer are in the tokenizer efficiency dataset. Rows marked ≈ are estimates from measured characters per token until you load that tokenizer; loading downloads its tokenizer.json from Hugging Face once and keeps it in your browser cache.

Create an account ($5 starting credit)

Other counters: DeepSeek · GLM · Qwen · Gemma · GPT · Llama 3 · Llama 4 · Kimi · Mistral Small · Ministral 3 · gpt-oss · Phi-4 · MiniMax · Nemotron 3 · Qwen3 · all models. BroadNet prices synced 2026-11-14.

Report a wrong price or number (email to info@broadnet.ai, with this page's link and your inputs filled in).

How it works

  1. Paste text into the box or pick a sample. The Gemma 3 tokenizer (Gemma 3 27B IT) is preselected and downloads once (33 MB) when you start typing.
  2. The text appears split into tokens, a colour per token; tick "Show token IDs" to see the numbers.
  3. Compare the Gemma 3 count with the other tokenizers in the table; load any of them for an exact count.
  4. Copy the link or download a CSV of the counts and costs.

About the Gemma 3 tokenizer

The Gemma 3 tokenizer is a SentencePiece-style BPE with byte fallback. It has 262,145 token IDs, 6,415 of them added or special tokens such as chat-format markers. Spaces become a ▁ marker that merges with the next word, and any byte not in the vocabulary falls back to a byte token, so no text is unknown. Numbers are split into single digits, so 2026 is four tokens.

On the open texts of our tokenizer efficiency dataset it needs 1,299 tokens per 1,000 English words (rank 11 of the 16 tokenizers we measured, 1 = fewest), 2,339 per 1,000 words of Python (rank 16 of 16) and 2,027 per 1,000 words of Modern Standard Arabic (rank 9 of 16). Its best showing here is Hindi (rank 2 of 16); its weakest is Python (rank 16 of 16).

3 models in our price database use this tokenizer: Gemma 3 27B, Gemma 3 12B and Gemma 3 4B. We hold 13 standard prices for them from 7 providers. The cheapest input price is $0.04 per 1M tokens, for Gemma 3 4B on Amazon Bedrock (checked 2026-11-14). At that price 1,000 English words cost $0.000052 to send. Gemma 3 models are not sold by BroadNet. That is less than BroadNet's cheapest chat model, Gemma 4 VL 26B, at $0.00013 for the same text.

Gemma 3 tokenizer facts
TokenizerGemma 3 tokenizer (Gemma 3 27B IT), a SentencePiece-style BPE with byte fallback
Sourcegoogle/gemma-3-27b-it (commit 005ad34), read from unsloth/gemma-3-27b-it (byte-identical)
Token IDs262,145 (262,144 in the base vocabulary)
Models using it3: Gemma 3 27B, Gemma 3 12B and Gemma 3 4B
Cheapest input price$0.04 per 1M tokens (Gemma 3 4B, Amazon Bedrock, checked 2026-11-14)
English prose130 tokens per 100 words (rank 11 of 16)
Arabic (MSA)203 tokens per 100 words (rank 9 of 16)
Python code234 tokens per 100 words (rank 16 of 16)
Download33 MB, once; then cached in your browser

Tokens per 100 words measured on the open texts of our dataset with all 16 tokenizers (2026-11-15); rank 1 = fewest tokens on the same texts. Model mapping checked 2026-11-15. Third-party prices are standard-tier prices from each provider's own page; BroadNet prices come from its pricing.json.

How much does a Gemma 3 prompt cost?

input cost  = tokens × input price per 1M  ÷ 1,000,000
output cost = tokens × output price per 1M ÷ 1,000,000

Worked example: "English Wikipedia: Water" in our dataset is 274 words. The Gemma 3 tokenizer (Gemma 3 27B IT) splits it into 356 tokens. Sent to Gemma 3 4B on Amazon Bedrock at $0.04 per 1M input tokens it costs $0.000014; the same number of tokens written as output at $0.08 per 1M costs $0.000028.

Questions

Which tokenizer does Gemma 3 use?

A SentencePiece-style BPE with byte fallback with 262,145 token IDs, published as tokenizer.json in google/gemma-3-27b-it. This page loads that file pinned to commit 005ad34, the commit our dataset measured. The official repo is gated (it needs a Hugging Face account and licence acceptance), so the file is read from unsloth/gemma-3-27b-it. Our dataset checks that the mirror's file is byte-identical to the official one: same SHA-256 and the same Git object ID that the official repo lists.

Which models use the Gemma 3 tokenizer?

Gemma 3 27B, Gemma 3 12B and Gemma 3 4B. Each was checked against the model's own Hugging Face repo on 2026-11-15: same tokenizer file, or the same vocabulary and rules with identical counts on every dataset text.

What is the cheapest Gemma 3 API price per token?

Of the 13 standard prices we hold for Gemma 3 models, the lowest input price is $0.04 per 1M tokens (output $0.08), for Gemma 3 4B on Amazon Bedrock, read from the provider's pricing page on 2026-11-14. Promotional prices are not ranked. Prices differ by model size and provider: each model's price page lists every provider we hold.

How many tokens does Arabic take with Gemma 3?

Modern Standard Arabic sentences take 1.47× the tokens of their English translations with this tokenizer (rank 9 of 16 on the same texts). Hindi takes 1.31× and Chinese 1.09×. Measured on parallel sentences in our open dataset (2026-11-15). Paste your own text above, or use the Arabic token counter.

Is the Gemma 3 tokenizer the same as Gemma 4 VL 26B?

They gave identical counts on all 52 texts of our dataset, but they are separate tokenizers with different special tokens, so text that contains control markers can count differently. Load both in the table above to compare your own text. Gemma 4 VL 26B has its own counter.

Does the count include chat-template tokens?

No. For Gemma 3 this page counts your text only, as our dataset does. A chat request also adds the chat template's role markers and special tokens, so the billed prompt is a little longer than your text. Read usage.prompt_tokens on a real response to see the exact figure.

Go deeper

Read as MarkdownAsk ChatGPTAsk ClaudeAsk Perplexity