- Intelligence
- 9.9
- Rank 427 of 744 measured models
- Output speed
- 203 t/s
- Rank 35 of 403 measured models
- Price/1M
- $0.85
- Rank 349 of 659 measured models
- Context window
- 1M
- Rank 84 of 1199 measured models
Gemini 2.5 Flash (Non-reasoning) scores 9.9 on the Artificial Analysis Intelligence Index, ranking 427th of 744 measured models. The closest models on this index are K2-V2 (9.9), North Mini Code (9.9), and Qwen3 VL 235B A22B Instruct (9.9). At $0.85 per million tokens (3:1 input/output blended price), it is more expensive than Qwen3 VL 235B A22B Instruct ($0.509). With an output speed of 203 t/s, it is faster than North Mini Code (82 t/s) and Qwen3 VL 235B A22B Instruct (28.5 t/s).
Additional benchmarks7
Design Arena · 3d47.4%
Design Arena · Codecategories46.9%
Design Arena · Dataviz49.1%
Design Arena · Gamedev44.3%
Design Arena · Svg43.1%
Design Arena · Uicomponent48.9%
Design Arena · Website47.1%
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.
- MiniMax M1 40k
MiniMax · — · —
10.0+0.1——Compare - K2-V2
Institute of Foundation Models · — · —
9.9——Compare - North Mini Code
Cohere · — · 82 t/s
9.9—82 t/sCompare - Qwen3 VL 235B A22B Instruct
Alibaba · $0.509 · 28.5 t/s
9.9$0.50928.5 t/sCompare
