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

Qwen

Qwen3.5 4B

Position

#120

Index

39.7

Estimated

publisher-corpus-prior over 197 held-out anchors (28%); launch card against 3 resolved peers on 10 rows (72%); exceeds every named peer on 2 of 10 rows; optimism haircut 0.6 from cross-lab-probe; shrunk 4.3 up toward the measured corpus

Comparison side 2

Output / 1M

Free

vs Unavailable

Context

256K

vs

Weights

Open weights

vs Open weights