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
TeksGambarVideo
Harga in / out
₩114 / ₩456
1M
Context
1 jt
Dirilis
26 Feb 2026

Providers

Beberapa tempat menghosting model yang sama. Permintaan disebarkan berdasarkan prioritas dan kesehatan, dan local GPU selalu didahulukan jika tersedia.

ProviderContextQuantizationMax outputPriorityState
OpenRouter1 jt50unknown