Multiverse Computing

Quasar 438B (Max, Based on GLM-5.2)

Aug 10, 2026ReasoningClosed weights

Compare

Intelligence

26.7

#156 of 748Rank 156 of 748 measured models

Coding

61.2

#81 of 298Rank 81 of 298 measured models

Output speed

172 t/s

#69 of 409Rank 69 of 409 measured models

Price/1M

$0.90

#374 of 670Rank 374 of 670 measured models

Context window

1M

#86 of 1215Rank 86 of 1215 measured models

Cost per task

$2.02

#169 of 197Rank 169 of 197 measured models

Quasar 438B (Max, Based on GLM-5.2) scores 26.7 on the Artificial Analysis Intelligence Index, ranking 156th of 748 measured models. The closest models on this index are Agnes 2.5 Pro Alpha (based on Qwen3.5 397B A17B) (26.8), GLM-5-Turbo (26.6), and Solar Pro 4 (26.6). At $0.90 per million tokens (3:1 input/output blended price), it is more expensive than Agnes 2.5 Pro Alpha (based on Qwen3.5 397B A17B) ($0.563) and Solar Pro 4 ($0.525). With an output speed of 172 t/s, it is faster than Agnes 2.5 Pro Alpha (based on Qwen3.5 397B A17B) (160 t/s) and Solar Pro 4 (118 t/s).