ModelCap

Current AI model comparison

Exact identities · public evidence

Microsoft

Fara1.5 27B

Position

#76

Index

51.2

Inherited

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

Comparison side 1

Poolside

Laguna XS 2.1

Position

#75

Index

51.9

Estimated

3 reported rows against measured corpus ladders (28%); global-corpus-prior over 185 held-out anchors (72%); shrunk 4.1 up toward the measured corpus

Comparison side 2

Output / 1M

Unavailable

vs $0.12

Context

—

vs 262K

Weights

Open weights

vs Restricted license