Kimi token counter
Updated
In short: This counter runs Moonshot AI's official Kimi tiktoken.model rank file in your browser, pinned to the same commit as our tokenizer dataset. It has 163,840 token IDs and is a 2.8 MB download. English prose takes about 129 tokens per 100 words, Modern Standard Arabic about 283.
tokens
| Model | Tokens | Chars per token | As input | As output |
|---|---|---|---|---|
| Kimi K2.6 Kimi tiktoken tokenizer (Kimi K3) Price: Kimi K2.6 at Chutes, 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). Kimi 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; Kimi 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.
Other counters: DeepSeek · GLM · Qwen · Gemma · GPT · Llama 3 · Llama 4 · Mistral Small · Ministral 3 · gpt-oss · Phi-4 · Gemma 3 · 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
- Paste text into the box or pick a sample. The Kimi tiktoken tokenizer (Kimi K3) is preselected and downloads once (2.8 MB) when you start typing.
- The text appears split into tokens, a colour per token; tick "Show token IDs" to see the numbers.
- Compare the Kimi count with the other tokenizers in the table; load any of them for an exact count.
- Copy the link or download a CSV of the counts and costs.
About the Kimi tokenizer
The Kimi tokenizer is a tiktoken byte-pair encoding (BPE). It has 163,840 token IDs, 256 of them added or special tokens such as chat-format markers. Moonshot publishes the rank file and the split rules (tokenization_kimi.py) instead of a tokenizer.json; runs of Chinese characters are split off before merging. Numbers are split into groups of up to three digits, so 2026 is two tokens.
On the open texts of our tokenizer efficiency dataset it needs 1,292 tokens per 1,000 English words (rank 9 of the 16 tokenizers we measured, 1 = fewest), 2,050 per 1,000 words of Python (rank 8 of 16) and 2,830 per 1,000 words of Modern Standard Arabic (rank 15 of 16). Its best showing here is Chinese (rank 1 of 16); its weakest is Arabic MSA (rank 15 of 16).
8 models in our price database use this tokenizer: Kimi K3, Kimi K3.1 Pro, Kimi K2.7 Code, Kimi K2.6, Kimi K2.5, Kimi K2 Thinking and 2 more. We hold 74 standard prices for them from 23 providers. The cheapest input price is $0.39 per 1M tokens, for Kimi K2.6 on Chutes (checked 2026-11-15). At that price 1,000 English words cost $0.00050 to send. Kimi models are not sold by BroadNet. BroadNet's cheapest chat model, Gemma 4 VL 26B, costs $0.00013 for the same text.
| Tokenizer | Kimi tiktoken tokenizer (Kimi K3), a tiktoken byte-pair encoding (BPE) |
|---|---|
| Source | moonshotai/Kimi-K3 (commit f831ab6) |
| Token IDs | 163,840 (163,584 in the base vocabulary) |
| Models using it | 8: Kimi K3, Kimi K3.1 Pro, Kimi K2.7 Code, Kimi K2.6, Kimi K2.5, Kimi K2 Thinking and 2 more |
| Cheapest input price | $0.39 per 1M tokens (Kimi K2.6, Chutes, checked 2026-11-15) |
| English prose | 129 tokens per 100 words (rank 9 of 16) |
| Arabic (MSA) | 283 tokens per 100 words (rank 15 of 16) |
| Python code | 205 tokens per 100 words (rank 8 of 16) |
| Download | 2.8 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 Kimi prompt cost?
input cost = tokens × input price per 1M ÷ 1,000,000
output cost = tokens × output price per 1M ÷ 1,000,000Worked example: "English Wikipedia: Water" in our dataset is 274 words. The Kimi tiktoken tokenizer (Kimi K3) splits it into 362 tokens. Sent to Kimi K2.6 on Chutes at $0.39 per 1M input tokens it costs $0.00014; the same number of tokens written as output at $1.80 per 1M costs $0.00065.
Questions
Which tokenizer does Kimi use?
A tiktoken BPE with 163,840 token IDs, published as tiktoken.model rank file in moonshotai/Kimi-K3. This page loads that file pinned to commit f831ab6, the commit our dataset measured.
Which models use the Kimi tokenizer?
Kimi K3, Kimi K3.1 Pro, Kimi K2.7 Code, Kimi K2.6, Kimi K2.5, Kimi K2 Thinking, Kimi K2 Instruct 0905 and Kimi K2 Instruct. 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 Kimi API price per token?
Of the 74 standard prices we hold for Kimi models, the lowest input price is $0.39 per 1M tokens (output $1.80), for Kimi K2.6 on Chutes, read from the provider's pricing page on 2026-11-15. 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 Kimi?
Modern Standard Arabic sentences take 2.05× the tokens of their English translations with this tokenizer (rank 15 of 16 on the same texts). Hindi takes 2.75× and Chinese 0.88×. Measured on parallel sentences in our open dataset (2026-11-15). Paste your own text above, or use the Arabic token counter.
Is the Kimi tokenizer the same as GPT (o200k)?
No. On the 52 texts of our dataset they gave the same count on 2; in total Kimi needs 35.5% more tokens, and up to 86% on a single text. See the GPT (o200k) counter.
Does the count include chat-template tokens?
No. For Kimi 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.