ModelCap
Current AI model comparison
Exact identities · public evidence
Qwen
Qwen3 4B Instruct 2507
Position
#151
Index
35.4
Estimated
publisher-corpus-prior over 203 held-out anchors (18%); launch card against 2 resolved peers on 9 rows (82%); exceeds every named peer on 7 of 9 rows; no cross-lab optimism probe available; shrunk 4 up toward the measured corpus
Comparison side 1
Tiiuae
Falcon H1R 7B
Position
#154
Index
34.3
Estimated
global-corpus-prior over 203 held-out anchors (8%); launch card against 2 resolved peers on 7 rows (92%); exceeds every named peer on 1 of 7 rows; optimism haircut 2.2 from cross-lab-probe; shrunk 1.9 up toward the measured corpus
Comparison side 2
Output / 1M
Unavailable
vs Unavailable
Context
262K
vs —
Weights
Open weights
vs Restricted license