ModelCap

Current AI model comparison

Exact identities · public evidence

Cohere

North Mini Code

Position

#122

Index

39.2

Measured

1 public benchmark observation across 1 board

Comparison side 1

Tiiuae

Falcon H1 34B Instruct

Position

#123

Index

39.2

Estimated

global-corpus-prior over 197 held-out anchors (47%); launch card against 5 resolved peers on 11 rows (53%); exceeds every named peer on 3 of 11 rows; optimism haircut 0 from cross-lab-probe; shrunk 10.8 up toward the measured corpus

Comparison side 2

Output / 1M

Free

vs Unavailable

Context

256K

vs

Weights

Open weights

vs Restricted license