# Alibaba Cloud Qwen3.8-Max API price

Alibaba Cloud Qwen3.8-Max: input $2.00, cached input $0.25, output $6.00 per 1M tokens (Standard, International (Singapore) scope).

- Input: $2.00 per 1M tokens
- Cached input: $0.25 per 1M tokens
- Output: $6.00 per 1M tokens
- Rate label: Standard, International (Singapore) scope
- Context window: 1M tokens
- Max output: 131,072 tokens
- Modalities: text, image, video → text
- Status: Active
- Release date: n/a
- API id: qwen3.8-max
- Weights from: Alibaba (Qwen)

International (Singapore) scope rate for prompts up to 1M tokens, thinking and non-thinking modes. Global and China (Beijing) scope list $1.65 input, $0.206 implicit-cache input, and $4.951 output per 1M tokens. Explicit cache creation is $2.50 and explicit cache reads $0.17 per 1M tokens on the International scope. Batch is 50% of real-time where supported and does not stack with cache discounts. The pricing page lists standard prices only; console promotions are not included. Context window 1M and max output 131,072 tokens are from the qwen3.8-max model page.

## Sources

- Pricing: https://www.alibabacloud.com/help/en/model-studio/model-pricing
- Limits: https://docs.modelstudio.console.alibabacloud.com/en/model-studio/qwen3-8-max

Checked: 2026-11-02

Canonical page: https://modelmeter.live/models/qwen3-8-max

Dataset: https://modelmeter.live/api/models.json

## Price changes

- 2026-11-02 new model: n/a -> $2.00 / $0.25 / $6.00 per 1M tokens (input / cached input / output). Alibaba Cloud lists Qwen3.8-Max (qwen3.8-max) at the International (Singapore) scope standard rate of $2 input, $0.25 implicit-cache input, and $6 output per 1M tokens for prompts up to 1M tokens; a release date is not published on the official page, and this row was added to the ledger on 2 November 2026. Source: https://www.alibabacloud.com/help/en/model-studio/model-pricing