ModelCap

Current AI model comparison

Exact identities · public evidence

Poolside

Laguna XS 2.1

Position

#86

Index

50.2

Estimated

3 reported rows against measured corpus ladders (29%); global-corpus-prior over 197 held-out anchors (71%); shrunk 2.8 up toward the measured corpus

Comparison side 1

Qwen

Qwen3 VL 235B A22B Thinking

Position

#87

Index

49.7

Measured

2 public benchmark observations across 2 boards

Comparison side 2

Output / 1M

$0.12

vs $4.00

Context

262K

vs 131K

Weights

Restricted license

vs Open weights