- Intelligence
- 35.5
- Rank 76 of 781 measured models
- Output speed
- 64.3 t/s
- Rank 316 of 449 measured models
- Price/1M
- $3
- Rank 549 of 740 measured models
- Context window
- 500K
- Rank 389 of 1323 measured models
- Cost per task
- $7.77
- Rank 218 of 229 measured models
Grok 4.7 (Xhigh) scores 35.5 on the Artificial Analysis Intelligence Index, ranking 76th of 781 measured models. The closest models on this index are Muse Ember (36.0), Agnes 3.0 Flash (35.5), and Agnes 2.5 Pro Beta (35.2). At $3 per million tokens (3:1 input/output blended price), it is more expensive than Agnes 3.0 Flash ($0.075) and Agnes 2.5 Pro Beta ($0.15). With an output speed of 64.3 t/s, it is slower than Muse Ember (446 t/s) and Agnes 2.5 Pro Beta (121 t/s).
Additional benchmarks11
Design Arena · 3d47.8%
Design Arena · Blender3d60.0%
Design Arena · Codecategories47.7%
Design Arena · Codetovideo42.8%
Design Arena · Dataviz45.1%
Design Arena · Gamedev50.9%
Design Arena · Mobileapps54.1%
Design Arena · Pdf48.8%
Design Arena · Svg50.6%
Design Arena · Uicomponent49.1%
Design Arena · Website45.9%
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.
- Muse Ember
Meta · — · 446 t/s
36.0+0.5—446 t/sCompare - Agnes 3.0 Flash
Sapiens AI · $0.075 · —
35.5$0.075—Compare - Agnes 2.5 Pro Beta
Sapiens AI · $0.15 · 121 t/s
35.2−0.3$0.15121 t/sCompare - DeepSeek V4 Flash Vision (Max)
DeepSeek · $0.66 · 140 t/s
34.8−0.7$0.66140 t/sCompare
