ModelCap

Current AI model comparison

Exact identities · public evidence

Cohere

North Mini Code

Position

#138

Index

32.9

Estimated

3 reported rows against measured corpus ladders (56%); publisher-corpus-prior over 186 held-out anchors (44%); shrunk 13.7 toward the measured corpus

Comparison side 1

OpenAI

gpt oss safeguard 120b

Position

#141

Index

31.9

Inherited

finetune of openai/gpt-oss-120b · fp4 (100%)

Comparison side 2

Output / 1M

Free

vs Unavailable

Context

256K

vs —

Weights

Open weights

vs Open weights