Modelocracy
Request a list

Compare two models

Pick any two of the 10 models to see how often their lists overlap, where they split, and which picks only one of them made.

Suggested matchups

Most alike

GPT-6 LunavsKimi K3

Agreement

77% agreement · same #1 on 46 of 64 lists

 Most different

Gemini 4 ArgonvsMiniMax M3

Agreement

67% agreement · same #1 on 35 of 64 lists

 Mainstream vs contrarian

Qwen 3.8 FlashvsKimi K3

Agreement

72% agreement · same #1 on 44 of 64 lists

 Most often the same #1

DeepSeek V4 ProvsMistral Large 4

Agreement

74% agreement · same #1 on 47 of 64 lists

Every pair

Mean overlap between two models' lists across every list both answered · 100 means identical · Color runs from 67 (lightest) to 77 (darkest) · Select a cell to compare that pair

Claude Haiku 5.5GPT-6 LunaGemini 4 ArgonQwen 3.8 FlashDeepSeek V4 ProGLM-5.3Kimi K3MiniMax M3Mistral Large 4Nemotron 3 Ultra
Claude Haiku 5.5737171707373707172
GPT-6 Luna737571747477707371
Gemini 4 Argon717567727375677068
Qwen 3.8 Flash717167697172687169
DeepSeek V4 Pro707472697576717472
GLM-5.3737473717576737373
Kimi K3737775727676717474
MiniMax M3707067687173717171
Mistral Large 4717370717473747173
Nemotron 3 Ultra727168697273747173