qwen3-vl-plus
On this page
Alibaba model qwen3-vl-plus. Prices are in US dollars per million tokens.
Current price, since 2026-10-09: input $0.20, output $1.60, cached input $0.04, cache write $0.25.
Matches Alibaba's pricing page through a correction, where models.dev differs, checked 2026-10-22 21:28 UTC.
Price a request with this model.
Standard price history#
| From | Input | Cached input | Cache write | Output | Longer prompts | Dated by | Source |
|---|---|---|---|---|---|---|---|
| 2026-10-09 | $0.20 | $0.04 | $0.25 | $1.60 | above 32K: in $0.30, cached $0.06, write $0.375, out $2.40; above 128K: in $0.60, cached $0.12, write $0.75, out $4.80 | documented | link |
| 2025-10-14 | $0.20 | $1.60 | above 32K: in $0.30, out $2.40; above 128K: in $0.60, out $4.80 | documented | link | ||
| 2025-10-14, first record | $0.20 | $1.60 | models.dev commit | link |
Thinking mode price history#
| From | Input | Cached input | Cache write | Output | Longer prompts | Dated by | Source |
|---|---|---|---|---|---|---|---|
| 2026-10-09 | $0.20 | $0.04 | $0.25 | $4.80 | documented | link | |
| 2025-10-14 | $0.20 | $4.80 | documented | link | |||
| 2025-10-14, first record | $0.20 | $4.80 | models.dev commit | link |
A blank cache price is charged at the input price. Longer prompts list the prices for requests whose whole prompt exceeds the size given.
import model_prices
model_prices.rates("qwen3-vl-plus", provider="alibaba")