ModelCap

Current AI model comparison

Exact identities · public evidence

DeepSeek

DeepSeek V4 Flash 0731

Position

#24

Index

70.7

Measured

2 public benchmark observations across 2 boards

Comparison side 1

Nex AGI

Nex-N2.5-Mini

Position

#22

Index

72.2

Estimated

global-corpus-prior over 182 held-out anchors (7%); launch card against 7 resolved peers on 8 rows (93%); optimism haircut 0 from cross-lab-probe; shrunk 1.4 toward the measured corpus

Comparison side 2

Output / 1M

$0.18

vs Free

Context

1.3M

vs 262K

Weights

Open weights

vs Open weights