Quasar 438B benchmarks, price, and specs | Tokens or Towers

AA Intelligence: 26.7#55 of 151
Input / output per 1M: $0.60/ $1.80
Output speed: 170tok/s
Context window: 1M tokens

Composite indexes

Artificial Analysis Intelligence Index#55 of 151
27independentIntelligence Index v4.3; max effortArtificial Analysis: Quasar 438B (max, based on GLM-5.2) ↗
Leader: Claude Opus 5.5 at 58

Coding

SciCode#52 of 100
48.1%independentmax effortArtificial Analysis: Quasar 438B (max, based on GLM-5.2) ↗
Leader: Claude Opus 5.5 at 66.9%
Terminal-Bench 4.0#57 of 96
1%independentmax effortArtificial Analysis: Quasar 438B (max, based on GLM-5.2) ↗
Leader: Claude Sonnet 5.5 at 70.6%
Terminal-Bench 2.x#38 of 105
69.3%independentTerminal-Bench 2.1; max effortArtificial Analysis: Quasar 438B (max, based on GLM-5.2) ↗
Leader: Claude Fable 5.1 at 91.4%

Agents and tool use

GDPval-AA#47 of 109
1162independentmax effortArtificial Analysis: Quasar 438B (max, based on GLM-5.2) ↗
Leader: Muse Wonder Full at 1857

Reasoning and knowledge

GPQA Diamond#97 of 144
73.2%independentmax effortArtificial Analysis: Quasar 438B (max, based on GLM-5.2) ↗
Leader: GPT-6 Astra at 96.1%
Humanity’s Last Exam#82 of 153
18.7%independentno tools; max effortArtificial Analysis: Quasar 438B (max, based on GLM-5.2) ↗
Leader: Claude Sonnet 5.5 at 64.5%
CritPt#46 of 144
9.4%independentmax effortArtificial Analysis: Quasar 438B (max, based on GLM-5.2) ↗
Leader: Muse Wonder Full at 33%

Vision and long context

AA Long Context Reasoning#60 of 144
76.3%independentmax effortArtificial Analysis: Quasar 438B (max, based on GLM-5.2) ↗
Leader: Kimi K3 at 88.7%

Other published results

Result	Score	Reported by	Source
tau2-Bench Banking (AA) · max effort	28.7%	independent	Artificial Analysis: Quasar 438B (max, based on GLM-5.2) ↗
AA-Omniscience Index · max effort	-2.6 index	independent	Artificial Analysis: Quasar 438B (max, based on GLM-5.2) ↗