qwen3.5-omni-flash-realtime

Realtime
Multimodal
Audio
Voice
Multilingual
Fast

Standard & Paid Price

Standard Price
MeterStandard Price
Input price$0.5500 / 1M tokens
Completion price$3.3000 / 1M tokens
Audio input price$4.5000 / 1M tokens
Audio completion price$17.7000 / 1M tokens
Billing details

Description

Alibaba Qwen3.5-Omni-Flash-Realtime — the fast realtime tier of the Qwen3.5 omni-modal line, held open as a WebSocket session for full-duplex speech conversation. Understands text, image, audio, and audio-video; audio input in 60+ languages and speech output in 30+, with steerable voice, web search, complex function calling, and semantic barge-in. Reached over GET /v1/realtime. Audio input and audio-bearing output are billed on their own faces, well above the text rate.

Capabilities

Input modalities
text, image, audio, video
Output modalities
text, audio

API endpoints

  • GET /v1/realtime (realtime)