ModelCap

Current AI model comparison

Exact identities · public evidence

Tencent

Hy4 preview

Position

#28

Index

76.4

Measured

9 reported rows against measured corpus ladders (46%); publisher-corpus-prior over 209 held-out anchors (10%); opens above measured predecessor tencent/hy3 on same-family ordering evidence (36%); 1 specialist-board observation at 2.5% support (8%); shrunk 1 toward the measured corpus

Comparison side 1

Xiaomi

MiMo-V2.6-Flash

Position

#10

Index

84.4

Estimated

publisher-corpus-prior over 209 held-out anchors (8%); launch card against 5 resolved peers on 7 rows (92%); exceeds every named peer on 1 of 7 rows; no cross-lab optimism probe available; shrunk 0.7 toward the measured corpus

Comparison side 2

Output / 1M

$2.50

vs $0.28

Context

1M

vs 1M

Weights

Open weights

vs Open weights