The Qwen3.5 series 397B-A17B 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. It delivers...
by Qwen|262K context|$0.39/M input tokens|$2.34/M output tokens
Endpoints
Available providers for this model, with details on pricing, context limits, and real-time health metrics.
