- Intelligence
- 38.8
- Rank 75 of 744 measured models
- Coding
- 72.4
- Rank 40 of 298 measured models
- Output speed
- 58.2 t/s
- Rank 287 of 403 measured models
- Price/1M
- $3
- Rank 487 of 659 measured models
- Context window
- 500K
- Rank 349 of 1199 measured models
- Cost per task
- $0.881
- Rank 122 of 193 measured models
Grok 4.5 (High) scores 38.8 on the Artificial Analysis Intelligence Index, ranking 75th of 744 measured models. The closest models on this index are Gemini 3.7 Flash (39.1), GPT-5.4 (39.0), and GPT-5.5 (38.4). At $3 per million tokens (3:1 input/output blended price), it is cheaper than GPT-5.4 ($5.63) and GPT-5.5 ($11.25), more expensive than Gemini 3.7 Flash ($1.50). With an output speed of 58.2 t/s, it is slower than Gemini 3.7 Flash (299 t/s), GPT-5.4 (85.5 t/s), and GPT-5.5 (81.9 t/s).
Additional benchmarks16
Design Arena · 3d50.3%
Design Arena · Agenticgamedev55.9%
Design Arena · Androidnative63.9%
Design Arena · Asciiart54.4%
Design Arena · Codecategories50.9%
Design Arena · Dataviz49.8%
Design Arena · Fullstack60.7%
Design Arena · Gamedev49.1%
Design Arena · Godotgamedev58.2%
Design Arena · Htmlslides52.5%
Design Arena · Mobileapps52.9%
Design Arena · Python Pptxslides53.9%
Design Arena · Svg48.2%
Design Arena · Uicomponent51.4%
Design Arena · Webapps53.5%
Design Arena · Website52.5%
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.
- Gemini 3.7 Flash
Google · $1.50 · 299 t/s
39.1+0.3$1.50299 t/sCompare - GPT-5.4
OpenAI · $5.63 · 85.5 t/s
39.0+0.2$5.6385.5 t/sCompare - GPT-5.5
OpenAI · $11.25 · 81.9 t/s
38.4−0.4$11.2581.9 t/sCompare - Mistral Large 4 Preview
Mistral · $2.07 · 116 t/s
38.4−0.4$2.07116 t/sCompare
