ModelCap

Current AI model comparison

Exact identities · public evidence

InclusionAI

AReaL SEA 235B A22B

Position

#84

Index

48.1

Inherited

finetune of qwen/qwen3-235b-a22b-thinking-2507 · finetune (100%)

Comparison side 1

StepFun

Step 3.7 Flash

Position

#83

Index

48.2

Estimated

3 reported rows against measured corpus ladders (14%); global-corpus-prior over 183 held-out anchors (19%); measured predecessor stepfun/step-3.5-flash less the succession penalty (67%); shrunk 1.2 up toward the measured corpus

Comparison side 2

Output / 1M

Unavailable

vs $1.15

Context

vs 262K

Weights

Open weights

vs Open weights