YFarmX logoYFarmX

AI Desk · AI Fingerprinting · Compare

Fireworks: Ember-1 · MoonshotAI: Kimi K3

Identity similarity: highExact tokenizer agreement · 50/50Exact agreement across every measured string is one strong signal of shared tokenizer lineage. The published settings differ, so the second signal the methodology needs for very high is absent here; the table below shows where they part.

Fireworks: Ember-1 recordMoonshotAI: Kimi K3 recordOpen in the interactive bench

Tokenizer fingerprint

50 of 50 mutually clean strings agree
Test stringFireworks: Ember-1MoonshotAI: Kimi K3Verdict
en-prose1414match
en-long1616match
spaces-2033match
spaces-6033match
tabs-2044match
newlines-2044match
mixed-ws66match
digits-933match
digits-1244match
digits-301010match
digits-sep77match
float-long99match
zh-common99match
zh-long1010match
zh-rare1818match
ja-kana1111match
ja-kanji1414match
ko1313match
ru1616match
ar1515match
he1616match
hi1717match
th2121match
el2222match
emoji-basic1010match
emoji-skin2020match
emoji-zwj-family1111match
emoji-zwj-x33333match
emoji-flags1414match
emoji-prof2020match
math2626match
boxdraw2424match
combining1414match
cjk-ext-b1616match
surrogates3131match
zalgo3636match
rtl-mix88match
py-code2525match
py-indent1515match
json2020match
html1616match
regex4343match
camel66match
snake66match
rare-word-x53131match
repeat-tok4141match
base643333match
hex1111match
url1717match
uuid2727match

Values are the marginal prompt-token cost of each string with the model's fixed overhead subtracted. An x marks a row excluded as corrupt for that model.

API surface

The declared serving contracts differ
FieldFireworks: Ember-1MoonshotAI: Kimi K3Verdict
Context length10485761048576✓ match
Max output tokens––✓ match
Supported parametersfrequency_penalty, include_reasoning, logit_bias, logprobs, max_tokens, presence_penalty, reasoning, reasoning_effort, repetition_penalty, response_format, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_pfrequency_penalty, include_reasoning, logit_bias, logprobs, max_tokens, min_p, presence_penalty, reasoning, reasoning_effort, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_p× differs
Defaults{}{"temperature":null,"top_p":0.95,"top_k":null,"frequency_penalty":null,"presence_penalty":null,"repetition_penalty":null}× differs
Reasoning contract{"mandatory":false,"default_enabled":true,"supported_efforts":["max","high","low"],"default_effort":"max"}{"mandatory":false,"default_enabled":true,"supported_efforts":["max","high","low"],"default_effort":"max"}✓ match
Modalitiestext+image->texttext+image+video->text× differs

Fireworks: Ember-1: OpenRouter's listing shows 943,718 output tokens, which is 90% of the context window, the figure it carries where the provider declares no maximum. MoonshotAI: Kimi K3: OpenRouter's listing shows 943,718 output tokens, which is 90% of the context window, the figure it carries where the provider declares no maximum.

Method: tokenizer fingerprinting, v1, 50 strings. This page is static and dated; the interactive bench compares any two of the 283 measured models.