ModelCap
Current AI model comparison
Exact identities · public evidence
Qwen
Qwen3 Coder 30B A3B Instruct
Position
#89
Index
48.1
Estimated
publisher-corpus-prior over 185 held-out anchors (7%); launch card against 3 resolved peers on 4 rows (94%); no cross-lab optimism probe available; shrunk 0.2 up toward the measured corpus
Comparison side 1
StepFun
Step 3.7 Flash
Position
#88
Index
48.3
Estimated
3 reported rows against measured corpus ladders (14%); global-corpus-prior over 185 held-out anchors (19%); measured predecessor stepfun/step-3.5-flash less the succession penalty (68%); shrunk 1.2 up toward the measured corpus
Comparison side 2
Output / 1M
$0.28
vs $1.15
Context
262K
vs 262K
Weights
Open weights
vs Open weights