MoonshotAI: Kimi K2 0711

moonshotai/kimi-k2 · Moonshot AINot yet established

Declared family: Unknown
Entered catalogue: 11 July 2025
Last tested: 22 August 2026
Measured signature: tk_2e353ea4

Identity statement

MoonshotAI: Kimi K2 0711 has a clean measured fingerprint that currently sits alone: no other measured model shares its signature.

What we checked

Four different kinds of evidence, and what each one showed

We checked	What we found	Where it came from
How it counts tokens	Counted cleanly, and no other model in the catalogue counts the same way	tk_2e353ea4
How its API is set up	A combination of settings no other listing in the catalogue uses	api_a64eec6d
How much it can read and write	Reads up to 131,072 tokens at once, writes up to 98,304	its own listing
Who served our requests	Novita	we measured it, 22 August 2026

Closest measured models

Agreement across the mutually clean test strings

MoonshotAI: Kimi K2.5moonshotai/kimi-k2.538/48 · 79%MoonshotAI: Kimi K3moonshotai/kimi-k339/50 · 78%MoonshotAI: Kimi K2.6moonshotai/kimi-k2.639/50 · 78%MoonshotAI: Kimi K2 Thinkingmoonshotai/kimi-k2-thinking39/50 · 78%MoonshotAI: Kimi K2.7 Codemoonshotai/kimi-k2.7-code39/50 · 78%Fireworks: Ember-1fireworks/ember-139/50 · 78%

Click a row to open the full comparison.

The measured fingerprint

What each test string cost this model, in tokens

We sent MoonshotAI: Kimi K2 0711 fifty short pieces of text and recorded what each one cost it in tokens. The bars below are those costs. Two models built on the same tokenizer produce the same bars; a model built on a different one produces a different set, which is what makes this a fingerprint.

All 50 rows

English and whitespace

en-prose13

en-long16

spaces-202

spaces-603

tabs-204

newlines-204

mixed-ws4

Digits

digits-93

digits-124

digits-3010

digits-sep7

float-long9

CJK

zh-common9

zh-long10

zh-rare16

ja-kana11

ja-kanji14

ko13

Other scripts

ru15

ar15

he16

hi17

th21

el22

Emoji

emoji-basic10

emoji-skin20

emoji-zwj-family9

emoji-zwj-x333

emoji-flags12

emoji-prof20

Rare Unicode

math26

boxdraw24

combining14

cjk-ext-b16

surrogates29

zalgo36

rtl-mix8

Code

py-code25

py-indent15

json20

html16

regex43

camel5

snake6

Repetition and encodings

rare-word-x531

repeat-tok40

base6433

hex11

url17

uuid25

Overhead subtracted: 9 prompt tokens.

Declared record

What the catalogue claims about this model

Kimi K2 Instruct is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass. It is optimized for...

Modalities: text->text
Declared tokenizer: Other
Prompt price: $0.57 / M tokens
Completion price: $2.30 / M tokens

Supported parameters

frequency_penaltymax_tokenspresence_penaltyrepetition_penaltyseedstoptemperaturetool_choicetoolstop_ktop_p

defaults: {}

Catalogue entry Weights on Hugging Face

History

Every observation, kept as taken

11 July 2025Enters the OpenRouter catalogue with no declared family.
22 August 2026Fingerprinted in the YFarmX catalogue sweep · 50 of 50 strings measured clean.