Qwen (Alibaba) API Pricing 2026
Alibaba's Qwen series via DashScope API. Qwen3.7 Max is the latest flagship; Qwen3 Coder 480B leads on coding tasks; Qwen3.5-Flash is among the cheapest capable models globally.
Pricing verified 2026-08-25. Sourced from www.alibabacloud.com/help/en/model-studio/model-pricing.
Get Qwen (Alibaba) API access →Qwen (Alibaba) Model Pricing
Prices in USD per 1M tokens
| Model | Input / 1M | Output / 1M | Context |
|---|---|---|---|
Qwen3.8 2.4T A95B Auto-added 2026-08; set isCurrent: true after review | $2 | $6 | 1,048,576 |
Qwen3.8 27B Auto-added 2026-08; set isCurrent: true after review | $0.43 | $2.55 | 1,000,000 |
Qwen3.8 Max Auto-added 2026-08; set isCurrent: true after review | $2 | $6 | 1,000,000 |
Qwen3.7 Plus Qwen3.7 Plus; 1M context; latest Alibaba generation | $0.32 | $1.28 | 1,000,000 |
Qwen3.7 Max Qwen3.7 Max; 1M context; top Alibaba flagship | $1.48 | $4.43 | 1,000,000 |
Qwen3.6 Flash Qwen3.6 Flash; 1M context; cheap tier | $0.19 | $1.13 | 1,000,000 |
Qwen3.6 Plus Qwen3.6 Plus; 1M context; mid-tier reasoning | $0.33 | $1.95 | 1,000,000 |
Qwen3.5-Flash Ultra-cheap high-volume tier; one of the lowest prices among capable models | $0.12 | $0.45 | 131,072 |
Qwen3.5-Plus Balanced tier; strong multilingual & coding; 128K context | $0.26 | $0.78 | 1,000,000 |
Qwen3 Coder Next Qwen3 Coder Next; latest code-specialized model | $0.12 | $0.8 | 262,144 |
Qwen3 Max Thinking Qwen3 Max with extended thinking mode | $0.78 | $3.9 | 262,144 |
Qwen3-Max Alibaba flagship; hybrid thinking mode; rivals GPT-4.1 at ~1/5 the cost | $0.45 | $1.82 | 131,072 |
Qwen3 VL 32B Instruct Auto-added 2026-06 | $0.1 | $0.42 | 131,072 |
Qwen3 Coder Plus Qwen3 Coder Plus; 1M context; strong code generation | $0.65 | $3.25 | 1,000,000 |
Qwen3 VL 235B A22B Instruct Auto-added 2026-06 | $0.21 | $1.9 | 262,144 |
Qwen3 Coder 480B A35B Qwen3 Coder 480B A35B; flagship code model; 1M context | $0.3 | $1 | 262,144 |
Qwen3 14B Auto-added 2026-06 | $0.12 | $0.24 | 131,072 |
Qwen3 30B A3B Auto-added 2026-06 | $0.12 | $0.5 | 131,072 |
Qwen2.5 VL 72B Instruct Auto-added 2026-06 | $0.8 | $1 | 128,000 |
Qwen2.5 Coder 32B Instruct Auto-added 2026-06 | $0.66 | $1 | 32,768 |
Qwen2.5 72B Instruct Auto-added 2026-06 | $0.36 | $0.4 | 32,768 |
How to read these estimates
The monthly table assumes 70% input tokens and 30% output tokens. It is a consistent comparison baseline, not a prediction of your workload. Actual bills can also include cached-input discounts, batch pricing, tool calls, minimum charges, regional differences, retries, and taxes.
The linked source is the provider's pricing documentation. Verify the current rate card and model availability before committing to a budget.
Estimated Monthly Cost (70% input / 30% output split)
| Model | 1M tokens/mo | 10M tokens/mo | 100M tokens/mo | 1B tokens/mo |
|---|---|---|---|---|
| Qwen3.8 2.4T A95B | $3.20 | $32.00 | $320 | $3,200 |
| Qwen3.8 27B | $1.07 | $10.66 | $107 | $1,066 |
| Qwen3.8 Max | $3.20 | $32.00 | $320 | $3,200 |
| Qwen3.7 Plus | $0.608 | $6.08 | $60.80 | $608 |
| Qwen3.7 Max | $2.37 | $23.65 | $236 | $2,365 |
| Qwen3.6 Flash | $0.472 | $4.72 | $47.20 | $472 |
| Qwen3.6 Plus | $0.816 | $8.16 | $81.60 | $816 |
| Qwen3.5-Flash | $0.219 | $2.19 | $21.90 | $219 |
| Qwen3.5-Plus | $0.416 | $4.16 | $41.60 | $416 |
| Qwen3 Coder Next | $0.324 | $3.24 | $32.40 | $324 |
| Qwen3 Max Thinking | $1.72 | $17.16 | $172 | $1,716 |
| Qwen3-Max | $0.861 | $8.61 | $86.10 | $861 |
| Qwen3 VL 32B Instruct | $0.196 | $1.96 | $19.60 | $196 |
| Qwen3 Coder Plus | $1.43 | $14.30 | $143 | $1,430 |
| Qwen3 VL 235B A22B Instruct | $0.717 | $7.17 | $71.70 | $717 |
| Qwen3 Coder 480B A35B | $0.510 | $5.10 | $51.00 | $510 |
| Qwen3 14B | $0.156 | $1.56 | $15.60 | $156 |
| Qwen3 30B A3B | $0.234 | $2.34 | $23.40 | $234 |
| Qwen2.5 VL 72B Instruct | $0.860 | $8.60 | $86.00 | $860 |
| Qwen2.5 Coder 32B Instruct | $0.762 | $7.62 | $76.20 | $762 |
| Qwen2.5 72B Instruct | $0.372 | $3.72 | $37.20 | $372 |
Frequently Asked Questions
How much does Qwen (Alibaba) LLM API cost?
Qwen (Alibaba) offers 21 models ranging from $0.100/1M to $2.00/1M input tokens. Alibaba's Qwen series via DashScope API. Qwen3.7 Max is the latest flagship; Qwen3 Coder 480B leads on coding tasks; Qwen3.5-Flash is among the cheapest capable models globally.
Is Qwen (Alibaba) cheaper than self-hosting?
For low-volume workloads (under 100M tokens/month), cloud APIs like Qwen (Alibaba) are almost always cheaper than purchasing and maintaining GPU hardware. Use our calculator to find the exact break-even point for your usage.