The Qwen3.5 122B-A10B native vision-language model is built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. In terms of...
Modalities
TextImageVideo
輸入 / 輸出價格
￦509 / ￦4,212
1M
Context
26.21萬
發佈日期
2026年2月26日
Pricing
價格以 KRW 為單位。匯率或利潤變動會立即反映。
| 輸入 (prompt) | ￦509 / 1M |
| 輸出 (completion) | ￦4,212 / 1M |
| Cache read | ￦51 / 1M |
| Cache write | — |
1,000 個輸入 tokens 加上 1,000 個輸出 tokens 的費用約為 ￦4.72。