- Intelligence
- 44.8
- Rank 38 of 744 measured models
- Coding
- 74.8
- Rank 30 of 298 measured models
- Output speed
- 61.1 t/s
- Rank 278 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
- $2.01
- Rank 163 of 193 measured models
GLM-5.3 (Max) scores 44.8 on the Artificial Analysis Intelligence Index, ranking 38th of 744 measured models. The closest models on this index are Qwen3.8 Max (0902) (45.4), Grok 4.6 (44.3), and Step 5 Preview (43.7). At $2.15 per million tokens (3:1 input/output blended price), it is cheaper than Qwen3.8 Max (0902) ($3) and Grok 4.6 ($3), more expensive than Step 5 Preview ($1.43). With an output speed of 61.1 t/s, it is faster than Qwen3.8 Max (0902) (42 t/s) and Grok 4.6 (50 t/s), slower than Step 5 Preview (89.3 t/s).
Additional benchmarks11
Design Arena · 3d59.5%
Design Arena · Asciiart47.2%
Design Arena · Codecategories54.2%
Design Arena · Dataviz51.7%
Design Arena · Gamedev52.5%
Design Arena · Htmlslides39.1%
Design Arena · Mobileapps52.2%
Design Arena · Python Pptxslides51.2%
Design Arena · Svg56.6%
Design Arena · Uicomponent57.0%
Design Arena · Website54.1%
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.
- Qwen3.8 Max (0902)
Alibaba · $3 · 42 t/s
45.4+0.6$342 t/sCompare - Grok 4.6
SpaceXSI · $3 · 50 t/s
44.3−0.5$350 t/sCompare - Step 5 Preview
StepFun · $1.43 · 89.3 t/s
43.7−1.1$1.4389.3 t/sCompare - Kimi K3
Moonshot AI · $6 · 43.3 t/s
43.6−1.2$643.3 t/sCompare
