ModelCap

Current AI model comparison

Exact identities · public evidence

NVIDIA

OpenMath2 Llama3.1 70B

Position

#204

Index

21.5

Inherited

finetune of meta-llama/Llama-3.1-70B · finetune (100%)

Comparison side 1

Tiiuae

Falcon H1 3B Instruct

Position

#201

Index

21.9

Estimated

global-corpus-prior over 209 held-out anchors (18%); launch card against 2 resolved peers on 11 rows (82%); exceeds every named peer on 9 of 11 rows; optimism haircut 10.03 from cross-lab-probe; shrunk 7.9 up toward the measured corpus

Comparison side 2

Output / 1M

Unavailable

vs Unavailable

Context

vs

Weights

Restricted license

vs Restricted license