Alibaba
Qwen3 Max (Preview)
ReasoningClosed weights
- Intelligence
- 16.3
- Rank 294 of 744 measured models
- Output speed
- 51.7 t/s
- Rank 313 of 403 measured models
- Price/1M
- $2.40
- Rank 480 of 659 measured models
- Context window
- 262K
- Rank 458 of 1199 measured models
Qwen3 Max Thinking (Preview) scores 16.3 on the Artificial Analysis Intelligence Index, ranking 294th of 744 measured models. The closest models on this index are Claude 4 Opus (16.6), Ring-2.6-1T (16.6), and Gemini 2.5 Pro (16.1). At $2.40 per million tokens (3:1 input/output blended price), it is cheaper than Claude 4 Opus ($30) and Gemini 2.5 Pro ($3.44), more expensive than Ring-2.6-1T ($0.85). With an output speed of 51.7 t/s, it is slower than Ring-2.6-1T (114 t/s) and Gemini 2.5 Pro (110 t/s).
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.
- Claude 4 Opus
Anthropic · $30 · —
16.6+0.3$30—Compare - Ring-2.6-1T
InclusionAI · $0.85 · 114 t/s
16.6+0.3$0.85114 t/sCompare - Step 3.5 Flash
StepFun · $0.15 · 138 t/s
16.6+0.3$0.15138 t/sCompare - Gemini 2.5 Pro
Google · $3.44 · 110 t/s
16.1−0.2$3.44110 t/sCompare
