Qwen: Qwen3.5-Flash
qwen/qwen3.5-flash-02-23
The Qwen3.5 native vision-language Flash models are built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. Compared to the...
Modalities
In / Out Price
$0.065 / $0.26 per 1M
Context
1M
Released
Feb 25, 2026
Knowledge Cutoff
—
Pricing
What Reroute charges per unit. No markup on inference — you pay the carrier's price, metered from the carrier's own usage numbers.
Input tokens
$0.065 / M
Output tokens
$0.26 / M
By provider
Each carrier sets its own price. The router weighs these against uptime when it picks one.
| Provider | Input /M | Output /M | Cache read /M | Cache write /M | Discount |
|---|---|---|---|---|---|
 Alibaba | $0.065 | $0.26 | — | — | — |