BroadRouter

Qwen3 token counter

Updated

In short: This counter runs Qwen's official Qwen3 tokenizer.json in your browser, pinned to the same commit as our tokenizer dataset. It has 151,669 token IDs and is an 11 MB download. English prose takes about 131 tokens per 100 words, Modern Standard Arabic about 229.

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
Qwen3 30B A3B (fp8)
Qwen3 tokenizer (151k vocabulary, Qwen3 32B)
Price: Qwen3 30B A3B (fp8) at Cloudflare Workers AI, 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). Qwen3 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; Qwen3 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 · Gemma 3 · MiniMax · Nemotron 3 · 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 Qwen3 tokenizer (151k vocabulary, Qwen3 32B) is preselected and downloads once (11 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 Qwen3 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 Qwen3 tokenizer

The Qwen3 tokenizer is a byte-level BPE. It has 151,669 token IDs, 26 of them added or special tokens such as chat-format markers, and applies NFC Unicode normalisation before splitting. Text is split into words, punctuation and numbers first, then merged byte by byte. Numbers are split into single digits, so 2026 is four tokens.

On the open texts of our tokenizer efficiency dataset it needs 1,306 tokens per 1,000 English words (rank 15 of the 16 tokenizers we measured, 1 = fewest), 2,018 per 1,000 words of Python (rank 4 of 16) and 2,285 per 1,000 words of Modern Standard Arabic (rank 13 of 16). Its best showing here is Python (rank 4 of 16); its weakest is English (rank 15 of 16).

17 models in our price database use this tokenizer: Qwen3 32B FP8, Qwen3-14B, Qwen3-8B, Qwen3 30B A3B (fp8), Qwen3-30B-A3B-Instruct-2507, Qwen3 235B A22B Instruct 2507 FP8 and 11 more. We hold 51 standard prices for them from 10 providers. The cheapest input price is $0.051 per 1M tokens, for Qwen3 30B A3B (fp8) on Cloudflare Workers AI (checked 2026-11-14). At that price 1,000 English words cost $0.000067 to send. Qwen3 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.

Qwen3 tokenizer facts
TokenizerQwen3 tokenizer (151k vocabulary, Qwen3 32B), a byte-level BPE
SourceQwen/Qwen3-32B (commit 9216db5)
Token IDs151,669 (151,643 in the base vocabulary)
Models using it17: Qwen3 32B FP8, Qwen3-14B, Qwen3-8B, Qwen3 30B A3B (fp8), Qwen3-30B-A3B-Instruct-2507, Qwen3 235B A22B Instruct 2507 FP8 and 11 more
Cheapest input price$0.051 per 1M tokens (Qwen3 30B A3B (fp8), Cloudflare Workers AI, checked 2026-11-14)
English prose131 tokens per 100 words (rank 15 of 16)
Arabic (MSA)229 tokens per 100 words (rank 13 of 16)
Python code202 tokens per 100 words (rank 4 of 16)
Download11 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 Qwen3 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 Qwen3 tokenizer (151k vocabulary, Qwen3 32B) splits it into 368 tokens. Sent to Qwen3 30B A3B (fp8) on Cloudflare Workers AI at $0.051 per 1M input tokens it costs $0.000019; the same number of tokens written as output at $0.335 per 1M costs $0.00012.

Questions

Which tokenizer does Qwen3 use?

A byte-level BPE with 151,669 token IDs, published as tokenizer.json in Qwen/Qwen3-32B. This page loads that file pinned to commit 9216db5, the commit our dataset measured.

Which models use the Qwen3 tokenizer?

Qwen3 32B FP8, Qwen3-14B, Qwen3-8B, Qwen3 30B A3B (fp8), Qwen3-30B-A3B-Instruct-2507, Qwen3 235B A22B Instruct 2507 FP8, Qwen3 Next 80B A3B Instruct, Qwen3 Next 80B A3B Thinking, Qwen3 Coder 480B A35B, Qwen3 Coder 30B A3B, Qwen3 Coder Next, Qwen3 VL 235B A22B, Qwen3 VL 30B A3B Instruct, Qwen3-VL-30B-A3B-Thinking, Qwen3-VL-32B-Instruct, Qwen3-VL-32B-Thinking and Qwen3-VL 8B. 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 Qwen3 API price per token?

Of the 51 standard prices we hold for Qwen3 models, the lowest input price is $0.051 per 1M tokens (output $0.335), for Qwen3 30B A3B (fp8) on Cloudflare Workers AI, 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 Qwen3?

Modern Standard Arabic sentences take 1.63× the tokens of their English translations with this tokenizer (rank 13 of 16 on the same texts). Hindi takes 4.43× and Chinese 1.01×. Measured on parallel sentences in our open dataset (2026-11-15). Paste your own text above, or use the Arabic token counter.

Is the Qwen3 tokenizer the same as Qwen3.8 27B?

No. On the 52 texts of our dataset they gave the same count on none of them; in total Qwen3 needs 36.6% more tokens, and up to 120% on a single text. See the Qwen3.8 27B counter.

Does the count include chat-template tokens?

No. For Qwen3 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