ModelCap
Current AI model comparison
Exact identities · public evidence
gemma 4 E4B it
Position
#102
Index
42.5
Estimated
publisher-corpus-prior over 186 held-out anchors (7%); launch card against 3 resolved peers on 5 rows (93%); no cross-lab optimism probe available; shrunk 1.6 up toward the measured corpus
Comparison side 1
Nanbeige
Nanbeige4.1 3B
Position
#99
Index
43.5
Estimated
global-corpus-prior over 186 held-out anchors (20%); launch card against 2 resolved peers on 6 rows (80%); exceeds every named peer on 5 of 6 rows; optimism haircut 3 from cross-lab-probe; shrunk 2.2 up toward the measured corpus
Comparison side 2
Output / 1M
Unavailable
vs Unavailable
Context
—
vs —
Weights
Open weights
vs Open weights