ModelCap

Current AI model comparison

Exact identities · public evidence

LGAI EXAONE

EXAONE 4.0.1 32B

Position

#122

Index

36.9

Estimated

global-corpus-prior over 186 held-out anchors (9%); launch card against 7 resolved peers on 8 rows (91%); optimism haircut 0 from cross-lab-probe; shrunk 1.5 up toward the measured corpus

Comparison side 1

Microsoft

Fara1.5 4B

Position

#119

Index

37.5

Inherited

finetune of Qwen/Qwen3.5-4B · finetune (100%)

Comparison side 2

Output / 1M

Unavailable

vs Unavailable

Context

vs

Weights

Restricted license

vs Open weights