ModelCap

Current AI model comparison

Exact identities · public evidence

Qwen

Qwen3 30B A3B Instruct 2507

Position

#100

Index

45.0

Measured

2 public benchmark observations across 2 boards

Comparison side 1

StepFun

Step 3.7 Flash

Position

#97

Index

47.7

Estimated

3 reported rows against measured corpus ladders (11%); global-corpus-prior over 194 held-out anchors (20%); measured predecessor stepfun/step-3.5-flash less the succession penalty (69%); shrunk 1.1 up toward the measured corpus

Comparison side 2

Output / 1M

$0.193

vs $1.15

Context

262K

vs 262K

Weights

Open weights

vs Open weights