ModelCap

Current AI model comparison

Exact identities · public evidence

Microsoft

Fara1.5 27B

Position

#72

Index

50.9

Inherited

finetune of qwen/qwen3.5-27b · finetune (100%)

Comparison side 1

Qwen

Qwen3.6 35B A3B

Position

#74

Index

50.7

Estimated

publisher-corpus-prior over 183 held-out anchors (13%); opens above measured predecessor qwen/qwen3.5-35b-a3b on same-family ordering evidence (46%); launch card against 4 resolved peers on 10 rows (41%); exceeds every named peer on 2 of 10 rows; optimism haircut 0 from cross-lab-probe; shrunk 0.1 toward the measured corpus

Comparison side 2

Output / 1M

Unavailable

vs $0.90

Context

vs 262K

Weights

Open weights

vs Open weights