- Intelligence
- 29.7
- Rank 130 of 744 measured models
- Coding
- 68.8
- Rank 59 of 298 measured models
- Output speed
- 131 t/s
- Rank 116 of 403 measured models
- Price/1M
- $4.50
- Rank 563 of 659 measured models
- Context window
- 1M
- Rank 84 of 1199 measured models
- Cost per task
- $1.30
- Rank 144 of 193 measured models
Gemini 3.1 Pro Preview scores 29.7 on the Artificial Analysis Intelligence Index, ranking 130th of 744 measured models. The closest models on this index are Claude Sonnet 4.6 (batch) (30.1), Qwen3.7 Max (29.5), and MiniMax-M3 (29.2). At $4.50 per million tokens (3:1 input/output blended price), it is more expensive than Qwen3.7 Max ($3.75) and MiniMax-M3 ($0.525). With an output speed of 131 t/s, it is faster than MiniMax-M3 (99.4 t/s), slower than Qwen3.7 Max (191 t/s).
Additional benchmarks21
Design Arena · 3d58.4%
Design Arena · Agenticgamedev44.2%
Design Arena · Agentichtmlslides55.8%
Design Arena · Agenticslides Html54.4%
Design Arena · Agenticslides Python Pptx33.9%
Design Arena · Agenticslides33.8%
Design Arena · Androidnative41.4%
Design Arena · Asciiart63.5%
Design Arena · Codecategories64.4%
Design Arena · Dataviz61.8%
Design Arena · Fullstack40.8%
Design Arena · Gamedev52.1%
Design Arena · Godotgamedev60.0%
Design Arena · Htmlslides48.6%
Design Arena · Mobileapps43.2%
Design Arena · Pptxslides34.1%
Design Arena · Python Pptxslides31.9%
Design Arena · Svg67.8%
Design Arena · Uicomponent69.4%
Design Arena · Webapps44.6%
Design Arena · Website64.4%
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.
- Claude Sonnet 4.6 (batch)
Anthropic · — · —
30.1+0.4——Compare - Qwen3.7 Max
Alibaba · $3.75 · 191 t/s
29.5−0.2$3.75191 t/sCompare - MiniMax-M3
MiniMax · $0.525 · 99.4 t/s
29.2−0.5$0.52599.4 t/sCompare - Claude Opus 4.5 (batch)
Anthropic · — · —
29.1−0.6——Compare
