The Qwen-Omni model accepts multimodal inputs including text, images, audio, and video, and generates text or speech responses. It offers a variety of expressive human-like voices and supports multilingual and dialect speech output, applicable to scenarios such as visual recognition, emotion sensing, and educational training. Usage \| API reference #### International If you select the International deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.
function_callinghttps://dashscope-intl.aliyuncs.com/compatible-mode/chat| Model | Context | Max out | Pricing |
|---|---|---|---|
| qwen3.5-122b-a10b | 256K | 64K | — |
| qwen3.5-27b | 256K | 64K | — |
| qwen3.5-35b-a3b | 256K | 64K | — |
| qwen3.5-397b-a17b | 256K | 64K | — |
| qwen3.5-livetranslate-flash-realtime | — | — | — |
| qwen3.5-ocr | — | — | — |
| qwen3.5-omni-flash-realtime | 262K | 66K | — |
| qwen3.5-omni-flash | 262K | — | — |
| qwen3.5-omni-plus-realtime | 262K | 66K | — |
| qwen3.5-omni-plus | 256K | — | — |
| qwen3.5-plus | 1M | 64K | — |
| qwen3.5-flash | 1M | 64K | $0.1$0.4 |
/v1/models/alibaba/qwen3.5-omni-flashThe Qwen-Omni model accepts multimodal inputs including text, images, audio, and video, and generates text or speech responses. It offers a variety of expressive human-like voices and supports multilingual and dialect speech output, applicable to scenarios such as visual recognition, emotion sensing, and educational training. Usage \| API reference #### International If you select the International deployment scope, model inference compute resources are dynamically scheduled worldwide, excluding the Chinese mainland. Static data is stored in your selected region. Supported region: Singapore.