Qwen pricing
Qwen API Cost Calculator
Estimate 通义千问 API cost in CNY for qwen-turbo, qwen-plus, qwen-max, and your own Alibaba Cloud Bailian contract rate. Compare monthly spend from real request volume, input tokens, and output tokens.
Qwen API cost calculator
| 厂商 | 模型 | 输入 / 百万 | 输出 / 百万 | 折扣 | 月成本 | 单次请求成本 |
|---|
Lowest listed cost
Start here for classification, support macros, routing, and other cost-sensitive workloads.
Balanced workloads
Use it when you need stronger answers but still want a low per-token operating cost.
Higher quality budget
Reserve it for workflows where answer quality matters more than pure token price.
When Qwen pricing analysis matters
Qwen API costs are sensitive to output length. A chatbot that writes short replies may fit qwen-turbo economics, while a report generator can shift most spend into output tokens. Use separate assumptions for support chat, RAG answers, agents, and content generation.
For a broader China model shortlist, compare Qwen against Doubao, ERNIE, and Hunyuan on the China LLM API Pricing Calculator. For a ByteDance-specific shortlist, use the Doubao API Cost Calculator.
| Cost lever | Why it matters for Qwen | What to check before launch |
|---|---|---|
| Output length | Output tokens usually cost more than input tokens. | Set max output limits for support, extraction, and report workflows separately. |
| Promotion price | Some public prices can be temporary or tied to resource packages. | Confirm whether your quote is list price, activity price, or contract price. |
| Repeated context | System prompts and RAG context can dominate monthly input tokens. | Ask whether cache, batch, or package pricing applies to your endpoint. |
FAQ
How do you estimate Qwen API cost?
Multiply monthly input tokens by the Qwen input price per million tokens, multiply monthly output tokens by the output price, then add the two costs. Use custom contract fields if your Alibaba Cloud quote differs from list pricing.
Which Qwen model is cheapest for high-volume API usage?
For the public prices tracked here, qwen-turbo has the lowest listed input and output token prices. qwen-plus and qwen-max may be useful when quality requirements justify a higher cost.
What can make a Qwen bill higher than this estimate?
Longer outputs, repeated context, retries, evaluation runs, batch settings, region differences, and expired promotion or resource-package pricing can all make the final bill differ from a simple token estimate.
Sources
Qwen prices were last checked on 2026-07-04. Verify the exact model endpoint in Alibaba Cloud before committing spend.