Qwen3 Max Thinking (Preview) API Pricing
Alibaba · released 2025-11-03
Pricing
0.120¢ / 1K
0.600¢ / 1K
0.240¢ / 1K
What it actually costs
| Workload | Tokens | Cost |
|---|---|---|
| Short chat reply | 500 in / 300 out | 0.24¢ |
| RAG query | 4,000 in / 500 out | 0.78¢ |
| Long document summary | 50,000 in / 1,000 out | $0.0660 |
| Code review of a 2k-line file | 30,000 in / 2,000 out | $0.0480 |
| 1,000 chat replies/day for 30 days | 30,000 × (500 in / 300 out) | $72.00 |
Cost = (input tokens ÷ 1,000,000 × $1.20) + (output tokens ÷ 1,000,000 × $6.00).
Benchmarks
#160 of 533
#51 of 249
Similar intelligence, lower price
| Model | Intelligence | Blended $/1M |
|---|---|---|
| Step 3.5 Flash 2603 | 17.0 | $0.15 |
| Step 3.5 Flash | 16.6 | $0.15 |
| Gemma 4 12B (Reasoning) | 14.2 | $0.15 |
| Qwen3.5 9B (Reasoning) | 13.7 | $0.151 |
| GLM-4.7-Flash (Reasoning) | 14.9 | $0.153 |
Leaderboard neighbours
| Model | Intelligence | Blended $/1M |
|---|---|---|
| DeepSeek V3.2 Exp (Reasoning) | 16.6 | $0.315 |
| Ring-2.6-1T | 16.6 | $0.85 |
| Step 3.5 Flash | 16.6 | $0.15 |
| Qwen3 Max Thinking (Preview) | 16.3 | $2.40 |
| Gemini 2.5 Pro | 16.1 | $3.438 |
| DeepSeek V3.2 (Non-reasoning) | 16.0 | $0.315 |
| MiMo-V2-Flash (Non-reasoning) | 16.0 | — |
More from Alibaba
| Model | Intelligence | Blended $/1M |
|---|---|---|
| Qwen3.8 Max (0902) | 45.4 | $3.00 |
| Qwen3.8 Max | 40.2 | $3.00 |
| Qwen3.8 2.4T A95B | 39.9 | $3.00 |
| Qwen3.8-Flash-Next | 39.8 | $0.23 |
| Qwen3.8 27B (xhigh) | 33.7 | $1.125 |
| Qwen3.7 Max | 29.5 | $3.75 |
| Qwen3.6 Max Preview | 28.4 | $2.925 |
| Qwen3.6 Plus | 27.0 | $1.125 |
Count tokens & estimate cost for Qwen3 Max Thinking (Preview) See Qwen3 Max Thinking (Preview) on the leaderboard
Source: Artificial Analysis · last synced 2026-09-22
FAQ
Who develops Qwen3 Max Thinking (Preview)?
Qwen3 Max Thinking (Preview) is developed by Alibaba, released 2025-11-03.
How much does Qwen3 Max Thinking (Preview) cost per million tokens?
Qwen3 Max Thinking (Preview) costs $1.20 per 1M input tokens and $6.00 per 1M output tokens ($2.40 blended), per Artificial Analysis's official pricing data.
How does Qwen3 Max Thinking (Preview) rank on benchmarks?
Qwen3 Max Thinking (Preview) scores 16.3 on the Artificial Analysis Intelligence Index, ranking #160 of 533 models we track.
Is there a cheaper model with similar intelligence to Qwen3 Max Thinking (Preview)?
Yes — Step 3.5 Flash 2603 scores similar intelligence (17.0) at $0.15/1M blended, versus $2.40/1M for Qwen3 Max Thinking (Preview).
Where does this data come from?
Pricing and benchmark data for Qwen3 Max Thinking (Preview) comes from Artificial Analysis (artificialanalysis.ai), synced daily. Last synced 2026-09-22.