- Intelligence
- 18.2
- Rank 270 of 744 measured models
- Coding
- 41.9
- Rank 152 of 298 measured models
- Output speed
- 127 t/s
- Rank 122 of 403 measured models
- Price/1M
- $0.844
- Rank 347 of 659 measured models
- Context window
- 262K
- Rank 458 of 1199 measured models
- Cost per task
- $0.475
- Rank 101 of 193 measured models
Qwen3.6 35B A3B (Reasoning) scores 18.2 on the Artificial Analysis Intelligence Index, ranking 270th of 744 measured models. The closest models on this index are Claude 4.1 Opus (18.6), Qwen3.5 397B A17B (18.4), and Gemini 3 Flash Preview (17.9). At $0.844 per million tokens (3:1 input/output blended price), it is cheaper than Claude 4.1 Opus ($30), Qwen3.5 397B A17B ($1.35), and Gemini 3 Flash Preview ($1.13). With an output speed of 127 t/s, it is faster than Qwen3.5 397B A17B (55.9 t/s), slower than Gemini 3 Flash Preview (195 t/s).
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.
- Claude 4.1 Opus
Anthropic · $30 · —
18.6+0.4$30—Compare - KAT-Coder-Pro V1
KwaiKAT · — · —
18.6+0.4——Compare - Qwen3.5 397B A17B
Alibaba · $1.35 · 55.9 t/s
18.4+0.2$1.3555.9 t/sCompare - Gemini 3 Flash Preview
Google · $1.13 · 195 t/s
17.9−0.3$1.13195 t/sCompare
