- Intelligence
- 33.7
- Rank 103 of 744 measured models
- Coding
- 68.8
- Rank 59 of 298 measured models
- Output speed
- 81.1 t/s
- Rank 227 of 403 measured models
- Price/1M
- $2.15
- Rank 474 of 659 measured models
- Context window
- 1M
- Rank 84 of 1199 measured models
- Cost per task
- $1.47
- Rank 149 of 193 measured models
GLM-5.2 (Max) scores 33.7 on the Artificial Analysis Intelligence Index, ranking 103rd of 744 measured models. The closest models on this index are Muse Spark 1.1 (Xhigh) (33.7), Qwen3.8 27B (33.7), and Motif 3 (33.6). At $2.15 per million tokens (3:1 input/output blended price), it is more expensive than Muse Spark 1.1 (Xhigh) ($2) and Qwen3.8 27B ($1.13). With an output speed of 81.1 t/s, it is faster than Qwen3.8 27B (41.2 t/s), slower than Muse Spark 1.1 (Xhigh) (148 t/s).
Additional benchmarks16
Design Arena · 3d53.3%
Design Arena · Agenticgamedev49.0%
Design Arena · Androidnative53.2%
Design Arena · Asciiart46.0%
Design Arena · Codecategories50.4%
Design Arena · Dataviz52.0%
Design Arena · Fullstack60.5%
Design Arena · Gamedev49.5%
Design Arena · Godotgamedev40.1%
Design Arena · Htmlslides49.5%
Design Arena · Mobileapps48.1%
Design Arena · Python Pptxslides48.7%
Design Arena · Svg48.8%
Design Arena · Uicomponent55.1%
Design Arena · Webapps56.0%
Design Arena · Website49.6%
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.
- JT-4.1 Flash 236B A21B
China Mobile · — · —
33.9+0.2——Compare - Muse Spark 1.1 (Xhigh)
Meta · $2 · 148 t/s
33.7$2148 t/sCompare - Qwen3.8 27B
Alibaba · $1.13 · 41.2 t/s
33.7$1.1341.2 t/sCompare - Motif 3
Motif Technologies · — · —
33.6−0.1——Compare
