Data Explorer

Per-model self-identification rates across languages, grouped by vendor. 361 cross-vendor confusions detected.

Run summary
21 models · 10 languages
6,300
Answers
83.0%
Self-ID
5.7%
Cross-vendor
10.0%
Unknown
1.2%
Refused
Imitation balance

Per manufacturer: how often other models claim to be it (right) vs how often its own models claim to be someone else (left). Sorted by net — identity creditors at the top, debtors at the bottom. Tested manufacturers only.

OpenAI is imitated most with a net imitation balance of 188; LongCat imitates others most with a net of -150.

OpenAI
+188
Google
+47
Anthropic
+33
z-ai
+21
Mistral
+14
Moonshot
+6
Meta
+2
Sakana AI
0
inclusionAI
-3
StepFun
-3
xAI
-5
MiniMax
-30
Xiaomi
-44
DeepSeek
-111
LongCat
-150
imitates others ◀│▶ imitated by others
Strongest confusion pairs

The most likely directed mistakes — when a manufacturer's models claim to be a specific other manufacturer. Bar length = how often (P across all answers for the source).

The strongest cross-vendor confusion is DeepSeek claiming to be OpenAI 40.7% of the time.

DeepSeek→OpenAI
40.7%122/300
LongCat→OpenAI
13.3%40/300
LongCat→Google
12.3%37/300
MiniMax→OpenAI
8.0%24/300
LongCat→Anthropic
7.7%23/300
LongCat→z-ai
7.0%21/300
LongCat→DeepSeek
2.3%7/300
Xiaomi→Mistral
2.3%14/600
Xiaomi→Anthropic
1.2%7/600
LongCat→Qwen
1.0%3/300
xAI→Moonshot
1.0%3/300
Xiaomi→Google
1.0%6/600
StepFun→Google
1.0%3/300
Xiaomi→DeepSeek
0.8%5/600
LongCat→AWS
0.7%2/300
By language

Self-ID and refusal rates per language, plus the most common cross-vendor confusion at that language.

LanguageSelf-IDRefusedTop confusion
ko한국어74.9%4.8%OpenAI53.3%
ruРусский79.0%4.1%Google30.0%
ja日本語79.5%3.5%OpenAI83.3%
ptPortuguês80.2%0.0%OpenAI40.0%
deDeutsch80.8%0.0%OpenAI40.0%
frFrançais82.1%0.0%OpenAI43.3%
esEspañol83.2%0.0%OpenAI56.7%
enEnglish87.9%0.0%OpenAI73.3%
zh-Hant繁體中文89.4%0.0%Anthropic10.0%
zh-Hans简体中文93.5%0.0%OpenAI10.0%
Answer composition by language

How the same “Who are you?” question splits into correct self-ID vs. confusion vs. abstention — per language, worst self-ID first.

Self-IDCross-vendorUnknownRefused

Self-identification is lowest in 한국어 at 74.9% and highest in 简体中文 at 93.5%.

ko한국어
75
10
10
75%
ruРусский
79
13
79%
ja日本語
80
9
80%
ptPortuguês
80
15
80%
deDeutsch
81
14
81%
frFrançais
82
11
82%
esEspañol
83
11
83%
enEnglish
88
88%
zh-Hant繁體中文
89
89%
zh-Hans简体中文
93
93%
Language fragility

Per model, the span of self-ID rate across the 10 languages (worst-language • → best-language •). A wide span means the model's sense of identity depends heavily on the language it's asked in. Widest swing first.

mistral-large-4 swings most — from 16.7% self-ID in ko to 100.0% in es.

mistral-large-4
17%ko
100%es
longcat-2.5-preview
0%ja
83%zh-Hans
deepseek-v4.1-flash
0%ja
80%zh-Hans
ling-3.1-flash
0%ja
77%en
minimax-m3.1-flash
0%ja
67%en
mimo-v2.6-pro
63%ko
100%en
mimo-v2.6-flash
77%fr
100%en
grok-4.7
90%en
100%ja
step-5-preview
90%ru
100%en
glm-5.3-prime
93%ko
100%en
gpt-6.1-terra
97%ko
100%en
gpt-6-luna
97%ru
100%en
glm-5.3-flashx
97%ru
100%en
gpt-6.1-sol
—
100%all
claude-haiku-5.5
—
100%all
muse-wonder-1.0
—
100%all
claude-sonnet-5.5
—
100%all
claude-opus-5.5
—
100%all
gemini-4-argon
—
100%all
kimi-k3.1-pro
—
100%all
fugu-ultra-v2
—
100%all
worst languagemeanbest language
Abstention by manufacturer

The other failure mode — not answering wrong, but not answering: giving no identity (“unknown”) or refusing outright. Share of all answers.

Unknown (“I’m an AI”)Refused

inclusionAI abstains most, with 75.3% of answers giving no usable identity.

inclusionAI
75.3%
MiniMax
63.7%
DeepSeek
44.3%
LongCat
33.3%
Mistral
13.7%
xAI
2.7%
Xiaomi
0.5%
z-ai
0.5%
OpenAI
0.2%
Anthropic
0.0%
Meta
0.0%
Google
0.0%
StepFun
0.0%
Moonshot
0.0%
Sakana AI
0.0%
By vendor

Rollup per real vendor — model count, total answers, mean self-ID rate, and the most common cross-vendor confusion target.

VendorModelsAnswersSelf-IDTop confusion
OpenAI390099.8%—
Anthropic3900100.0%—
Xiaomi260092.0%Mistral2.3%
z-ai260099.5%—
DeepSeek130014.3%OpenAI40.7%
LongCat130016.7%OpenAI13.3%
inclusionAI130023.7%Ayerolabs0.3%
MiniMax130026.0%OpenAI8.0%
Mistral130086.3%—
xAI130095.7%Moonshot1.0%
StepFun130099.0%Google1.0%
Meta1300100.0%—
Google1300100.0%—
Moonshot1300100.0%—
Sakana AI1300100.0%—
Qwen000.0%—
AWS000.0%—
Microsoft000.0%—
Ayerolabs000.0%—
阶点000.0%—
Kiro AI000.0%—
Amazon Web Services000.0%—
Kiss000.0%—
Advocate000.0%—
Amazon000.0%—
Quora000.0%—
Miyada000.0%—
Assistant000.0%—
Lenforcer000.0%—
智多星000.0%—
Max000.0%—
Kiro000.0%—
MobilityMaterial000.0%—
장젠데리안000.0%—
Tencent000.0%—
TZi Lun Czi-indong000.0%—
Ford000.0%—
Zenith AI Corp000.0%—
Zoom000.0%—
Cohere000.0%—
CoreAI000.0%—
Milq AI000.0%—
Doubao000.0%—
Hixxy000.0%—
Anthropic
3 models

Self-Identification Rate

Modelenzh-Hanszh-HantjakoruesfrdeptOverall
Claude Haiku 5.5100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%
Claude Sonnet 5.5100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%
Claude Opus 5.5100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%

Cross-Vendor Confusions

Claude Haiku 5.5100.0% self

✓ Always self-identifies correctly

Claude Sonnet 5.5100.0% self

✓ Always self-identifies correctly

Claude Opus 5.5100.0% self

✓ Always self-identifies correctly

OpenAI
3 models

Self-Identification Rate

Modelenzh-Hanszh-HantjakoruesfrdeptOverall
GPT-6.1 Terra100.0%100.0%100.0%100.0%96.7%100.0%100.0%100.0%100.0%100.0%99.7%
GPT-6.1 Sol100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%
GPT-6 Luna100.0%100.0%100.0%100.0%100.0%96.7%100.0%100.0%100.0%100.0%99.7%

Cross-Vendor Confusions

GPT-6.1 Terra99.7% self

✓ Always self-identifies correctly

GPT-6.1 Sol100.0% self

✓ Always self-identifies correctly

GPT-6 Luna99.7% self

✓ Always self-identifies correctly

Xiaomi
2 models

Self-Identification Rate

Modelenzh-Hanszh-HantjakoruesfrdeptOverall
MiMo V2.6 Flash100.0%100.0%100.0%93.3%96.7%96.7%96.7%76.7%96.7%86.7%94.3%
MiMo V2.6 Pro100.0%100.0%100.0%90.0%63.3%80.0%93.3%80.0%96.7%93.3%89.7%

Cross-Vendor Confusions

MiMo V2.6 Flash94.3% self
Mistaken as:Mistral2.0%DeepSeek1.0%Google0.7%Microsoft0.7%Qwen0.3%Hixxy0.3%Meta0.3%
MiMo V2.6 Pro89.7% self
Mistaken as:Mistral2.7%Anthropic2.3%Google1.3%DeepSeek0.7%MiniMax0.3%Cohere0.3%Microsoft0.3%CoreAI0.3%Milq AI0.3%Moonshot0.3%OpenAI0.3%Doubao0.3%
z-ai
2 models

Self-Identification Rate

Modelenzh-Hanszh-HantjakoruesfrdeptOverall
GLM 5.3 Prime100.0%100.0%100.0%100.0%93.3%100.0%100.0%100.0%100.0%100.0%99.3%
GLM 5.3 FlashX100.0%100.0%100.0%100.0%100.0%96.7%100.0%100.0%100.0%100.0%99.7%

Cross-Vendor Confusions

GLM 5.3 Prime99.3% self

✓ Always self-identifies correctly

GLM 5.3 FlashX99.7% self

✓ Always self-identifies correctly

DeepSeek
1 model

Self-Identification Rate

Modelenzh-Hanszh-HantjakoruesfrdeptOverall
DeepSeek V4.1 Flash3.3%80.0%20.0%0.0%0.0%6.7%13.3%10.0%6.7%3.3%14.3%

Cross-Vendor Confusions

DeepSeek V4.1 Flash14.3% self
Mistaken as:OpenAI40.7%Anthropic0.7%
Google
1 model

Self-Identification Rate

Modelenzh-Hanszh-HantjakoruesfrdeptOverall
Gemini 4 Argon100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%

Cross-Vendor Confusions

Gemini 4 Argon100.0% self

✓ Always self-identifies correctly

inclusionAI
1 model

Self-Identification Rate

Modelenzh-Hanszh-HantjakoruesfrdeptOverall
Ling 3.1 Flash76.7%66.7%70.0%0.0%0.0%0.0%13.3%10.0%0.0%0.0%23.7%

Cross-Vendor Confusions

Ling 3.1 Flash23.7% self
Mistaken as:Ayerolabs0.3%Anthropic0.3%Tencent0.3%
LongCat
1 model

Self-Identification Rate

Modelenzh-Hanszh-HantjakoruesfrdeptOverall
LongCat 2.5 Preview16.7%83.3%50.0%0.0%0.0%0.0%0.0%16.7%0.0%0.0%16.7%

Cross-Vendor Confusions

LongCat 2.5 Preview16.7% self
Mistaken as:OpenAI13.3%Google12.3%Anthropic7.7%z-ai7.0%DeepSeek2.3%Qwen1.0%AWS0.7%Moonshot0.7%阶点0.3%Kiro AI0.3%Amazon Web Services0.3%Xiaomi0.3%Miyada0.3%Assistant0.3%智多星0.3%Max0.3%MobilityMaterial0.3%장젠데리안0.3%TZi Lun Czi-indong0.3%Ford0.3%Meta0.3%Zenith AI Corp0.3%Zoom0.3%
Meta
1 model

Self-Identification Rate

Modelenzh-Hanszh-HantjakoruesfrdeptOverall
Muse Wonder 1.0100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%

Cross-Vendor Confusions

Muse Wonder 1.0100.0% self

✓ Always self-identifies correctly

MiniMax
1 model

Self-Identification Rate

Modelenzh-Hanszh-HantjakoruesfrdeptOverall
MiniMax M3.1 Flash66.7%50.0%53.3%0.0%6.7%3.3%30.0%33.3%10.0%6.7%26.0%

Cross-Vendor Confusions

MiniMax M3.1 Flash26.0% self
Mistaken as:OpenAI8.0%Kiss0.3%Advocate0.3%Amazon0.3%Quora0.3%Lenforcer0.3%Google0.3%DeepSeek0.3%
Mistral
1 model

Self-Identification Rate

Modelenzh-Hanszh-HantjakoruesfrdeptOverall
Mistral Large 493.3%93.3%86.7%86.7%16.7%90.0%100.0%100.0%96.7%100.0%86.3%

Cross-Vendor Confusions

Mistral Large 486.3% self

✓ Always self-identifies correctly

Moonshot
1 model

Self-Identification Rate

Modelenzh-Hanszh-HantjakoruesfrdeptOverall
Kimi K3.1 Pro100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%

Cross-Vendor Confusions

Kimi K3.1 Pro100.0% self

✓ Always self-identifies correctly

Sakana AI
1 model

Self-Identification Rate

Modelenzh-Hanszh-HantjakoruesfrdeptOverall
Fugu Ultra v2100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%100.0%

Cross-Vendor Confusions

Fugu Ultra v2100.0% self

✓ Always self-identifies correctly

StepFun
1 model

Self-Identification Rate

Modelenzh-Hanszh-HantjakoruesfrdeptOverall
Step 5 Preview100.0%100.0%100.0%100.0%100.0%90.0%100.0%100.0%100.0%100.0%99.0%

Cross-Vendor Confusions

Step 5 Preview99.0% self
Mistaken as:Google1.0%
xAI
1 model

Self-Identification Rate

Modelenzh-Hanszh-HantjakoruesfrdeptOverall
Grok 4.790.0%90.0%96.7%100.0%100.0%100.0%100.0%96.7%90.0%93.3%95.7%

Cross-Vendor Confusions

Grok 4.795.7% self
Mistaken as:Moonshot1.0%OpenAI0.3%Kiro0.3%
v0.65.0 · af80a02-dirty ·