ModelCap

Current AI model comparison

Exact identities · public evidence

MiniMax

MiniMax-01

Position

#135

Index

42.0

Estimated

publisher-corpus-prior over 215 held-out anchors (44%); launch card against 2 resolved peers on 4 rows (56%); optimism haircut 6.45 from cross-lab-probe; shrunk 8.3 up toward the measured corpus

Comparison side 1

Qwen

Qwen Drive 1.0 4B

Position

#132

Index

44.1

Inherited

finetune of Qwen/Qwen3.5-4B · finetune (100%)

Comparison side 2

Output / 1M

$1.10

vs Unavailable

Context

1M

vs —

Weights

License unverified

vs Open weights