ModelCap

Current AI model comparison

Exact identities · public evidence

IFM

K2 Horizon 32B

Position

#97

Index

47.8

Estimated

global-corpus-prior over 197 held-out anchors (49%); launch card against 2 resolved peers on 8 rows (51%); exceeds every named peer on 1 of 8 rows; optimism haircut 1.3 from cross-lab-probe; shrunk 3.3 up toward the measured corpus

Comparison side 1

NVIDIA

Ising Calibration 1 35B A3B

Position

#95

Index

48.3

Inherited

finetune of qwen/qwen3.5-35b-a3b · finetune (100%)

Comparison side 2

Output / 1M

Unavailable

vs Unavailable

Context

vs

Weights

Open weights

vs Restricted license