Capable of debugging code, automating office workflows, and executing autonomously across long-horizon tasks.
Discount applies to all 4 billing items: Input, Output, Explicit Cache Creation, and Explicit Cache Hit.
New customers get 1M free tokens for Qwen3.7-Max and 1M for Qwen3.7-Plus.








Capable of debugging code, automating office workflows, and executing autonomously across long-horizon tasks.


Discount applies to all 4 billing items: Input, Output, Explicit Cache Creation, and Explicit Cache Hit.


New customers get 1M free tokens for Qwen3.7-Max and 1M for Qwen3.7-Plus.
Deep Thinking & Text Generation
List Price: $1.65 / 1M tokens
128K Long Context - Tool Use + Visual Input
Supports complex reasoning/ agent & production Q&A
Off-peak tokens, 80% off (22:00–08:00 UTC+8)
Deep Thinking & Text Generation
List Price: $0.276 / 1M tokens
128K Long Context - High QPS + Multi-modal
Supports function calls/ multi-turn dialogue/ code generation
Off-peak tokens, 60% off (22:00–08:00 UTC+8)

Q: How is the off-peak discount window determined?
A: Based on the actual call time (UTC+8): Any API call received by the server between 22:00 and 08:00 the next day will be billed at the off-peak rate. Bills are labeled "Off-Peak" for easy verification.
Q: Which models are covered? Does it include the latest versions?
A: The event covers the latest Qwen series models including Qwen 3.7-Max and Qwen 3.7-Plus.You get the same model version day and night — no off-peak downgrade
Q: Can the off-peak discount stack with Enterprise Credits or vouchers?
A: Yes. The off-peak discount is applied before Credit / voucher deductions in the billing process. Actual bill = Night discounted price × Usage – Available Credit deduction. Extra credits obtained through the Enterprise Credit Program can also be used during the event.
Q: Which regions are eligible for this discount?
A: Available in global mode in the following regions: Germany, Hong Kong (China), Japan, and US (Virginia).