ModelCap
Current AI model comparison
Exact identities · public evidence
Amazon
Nova Pro 1.0
Position
#187
Index
23.0
Measured
2 public benchmark observations across 2 boards
Comparison side 1
Microsoft
Phi 3.5 mini instruct
Position
#186
Index
24.1
Estimated
global-corpus-prior over 209 held-out anchors (26%); launch card against 4 resolved peers on 6 rows (74%); exceeds every named peer on 1 of 6 rows; optimism haircut 0.67 from cross-lab-probe; shrunk 11.5 up toward the measured corpus
Comparison side 2
Output / 1M
$3.20
vs Unavailable
Context
300K
vs —
Weights
API only
vs Open weights