Qwen: Qwen3 VL 32B Instruct
qwen/qwen3-vl-32b-instruct
Qwen3-VL-32B-Instruct is a large-scale multimodal vision-language model designed for high-precision understanding and reasoning across text, images, and video. With 32 billion parameters, it combines deep visual perception with advanced text...
Modalities
In / Out Price
$0.104 / $0.416 per 1M
Context
131K
Released
Oct 23, 2025
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
$0.104 / M
Output tokens
$0.416 / 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 | $0.104 | $0.416 | — | — | — |