Model board / Qwen3.8-Flash

Verified 2026-09-10 against the provider's own page. Prices are USD per 1M tokens, list rate — no committed-use, enterprise or volume discount.

Model Studio international rate, one tier up to 1M input tokens. The page flags a context-caching discount but prints no rate for it. Together lists the same $0.15/$0.47

not stated — the price tier runs to 1M input tokens, which is a billing band, not a window

a context-caching discount is flagged on the pricing page; no rate or mechanism is printed

Watch out

Priced, and little else: the page that prices it states no window, no cached rate and no capabilities. Treat it as unmeasured until you have run it

What we would use it for

Alibaba's cheapest current-generation Qwen tier

What this has cost, week by week

Output price at each refresh since 2026-09-10. A vendor change is the provider changing its price. Our correction means this board changed what it prints and the provider's charge did not move — the two are never merged, because publishing the second as the first would be a price rise that never happened.

Nothing has moved since we started tracking it.

Where these numbers came from

• https://www.alibabacloud.com/help/en/model-studio/model-pricing

No number on this page was written that was not read from one of those pages on 2026-09-10. If one is wrong, send us the source URL and it is corrected in the next weekly refresh, with a changelog line saying what was wrong — contact@naderu.com.

The rest of the board

This row sits alongside 57 others on the model board, refreshed every week. The same data as JSON, and the terms for using it, are at /board-data.