- Intelligence
- 30.4
- Rank 122 of 744 measured models
- Output speed
- 73.3 t/s
- Rank 245 of 403 measured models
- Price/1M
- $4.81
- Rank 574 of 659 measured models
- Context window
- 400K
- Rank 357 of 1199 measured models
GPT-5.2 (Xhigh) scores 30.4 on the Artificial Analysis Intelligence Index, ranking 122nd of 744 measured models. The closest models on this index are K2 Horizon 375B A23B (30.5), DeepSeek V4 Pro 0424 (30.4), and Claude Sonnet 4.6 (batch) (30.1). At $4.81 per million tokens (3:1 input/output blended price), it is more expensive than DeepSeek V4 Pro 0424 ($0.544). With an output speed of 73.3 t/s, it is slower than K2 Horizon 375B A23B (474 t/s) and DeepSeek V4 Pro 0424 (83.4 t/s).
Additional benchmarks13
Design Arena · 3d41.5%
Design Arena · Androidnative49.1%
Design Arena · Asciiart50.7%
Design Arena · Codecategories49.6%
Design Arena · Dataviz56.1%
Design Arena · Fullstack44.3%
Design Arena · Gamedev56.0%
Design Arena · Godotgamedev48.1%
Design Arena · Mobileapps46.8%
Design Arena · Svg53.7%
Design Arena · Uicomponent51.3%
Design Arena · Webapps45.5%
Design Arena · Website54.4%
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.
- K2 Horizon 375B A23B
Institute of Foundation Models · — · 474 t/s
30.5+0.1—474 t/sCompare - DeepSeek V4 Pro 0424
DeepSeek · $0.544 · 83.4 t/s
30.4$0.54483.4 t/sCompare - Claude Sonnet 4.6 (batch)
Anthropic · — · —
30.1−0.3——Compare - Gemini 3.1 Pro Preview
Google · $4.50 · 131 t/s
29.7−0.7$4.50131 t/sCompare
