Overview
Qwen3.5 Omni Flash is a fast multimodal model supporting text, audio, image, and video inputs with text and audio output. It enables real-time multimodal speech conversations with improved understanding over Qwen3 Omni Flash.
Key Features
- Real-time speech — Optimized for real-time multimodal speech conversation
- Omni-modal input — Accepts text, audio, image, and video inputs
- Text + audio output — Generate speech responses alongside text
- Improved accuracy — Better multimodal understanding than Qwen3 Omni Flash
- Affordable — Text/image/video input from $0.40 per 1M tokens
- OpenAI-compatible API — Drop-in replacement using chat completions format
Pricing
Per Million Tokens
Text Input: $0.40
Audio Input: $3.00
Image/Video Input: $0.40
Text Only Output: $2.20
Text Multimodal Output: $2.20
Audio Output: $11.90
Parameters
No input schema available.