ModelCap
Current AI model comparison
Exact identities · public evidence
IFM
K2 Horizon MoVA 36B A4B
Position
#78
Index
54.5
Estimated
global-corpus-prior over 195 held-out anchors (77%); launch card against 3 resolved peers on 8 rows (23%); exceeds every named peer on 4 of 8 rows; optimism haircut 2.24 from cross-lab-probe; shrunk 6 toward the measured corpus
Comparison side 1
Microsoft
Fara1.5 27B
Position
#81
Index
52.0
Inherited
finetune of qwen/qwen3.5-27b · finetune (100%)
Comparison side 2
Output / 1M
Unavailable
vs Unavailable
Context
—
vs —
Weights
Open weights
vs Open weights