- Intelligence
- 18.4
- Rank 268 of 744 measured models
- Coding
- 48.2
- Rank 129 of 298 measured models
- Output speed
- 55.9 t/s
- Rank 298 of 403 measured models
- Price/1M
- $1.35
- Rank 418 of 659 measured models
- Context window
- 262K
- Rank 458 of 1199 measured models
- Cost per task
- $0.475
- Rank 100 of 193 measured models
Qwen3.5 397B A17B (Reasoning) scores 18.4 on the Artificial Analysis Intelligence Index, ranking 268th of 744 measured models. The closest models on this index are Claude 4.1 Opus (18.6), KAT-Coder-Pro V1 (18.6), and Qwen3.6 35B A3B (18.2). At $1.35 per million tokens (3:1 input/output blended price), it is cheaper than Claude 4.1 Opus ($30), more expensive than Qwen3.6 35B A3B ($0.844). With an output speed of 55.9 t/s, it is slower than Qwen3.6 35B A3B (127 t/s).
Additional benchmarks7
Design Arena · 3d56.7%
Design Arena · Codecategories52.6%
Design Arena · Dataviz53.2%
Design Arena · Gamedev50.2%
Design Arena · Svg56.1%
Design Arena · Uicomponent51.4%
Design Arena · Website52.5%
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.
- Claude 4.1 Opus
Anthropic · $30 · —
18.6+0.2$30—Compare - KAT-Coder-Pro V1
KwaiKAT · — · —
18.6+0.2——Compare - MiniMax-M2
MiniMax · $0.525 · 64.7 t/s
18.6+0.2$0.52564.7 t/sCompare - Qwen3.6 35B A3B
Alibaba · $0.844 · 127 t/s
18.2−0.2$0.844127 t/sCompare
