QwenMeta🚀 Qwen3.5-Omni is Live! Plus & Flash variants now on Model Studio 🔹 Native Omni-Modal: Text + Image + Audio + Video in one model 🔹 Thinker-Talker Architecture: Hybrid MoE + 256K context → faster & smarter 🔹 Long-Form Input: Up to 3h audio / 1h video (non-realtime API) 🔹 113 Languages: Speech recognition + 36 for synthesis 🔹 SOTA Results: Tops 215 subtasks, outperforms Gemini-3.1 Pro 🔹 Real-Time Ready: Semantic interruption, WebSearch, FunctionCall, voice control & cloning 🔹 ARIA Tech: Smoother streaming TTS, no skips or misreads 📦 Models: qwen3.5-omni-plus | plus-realtime (high perf) qwen3.5-omni-flash | flash-realtime (high speed) → All with 256K context 🔗 Quick Links: • 🧪 Try: https://chat.qwen.ai/ • 💻 Demo Code: https://code.alibaba-inc.com/qwen-demo/qwen-omni-web • 📚 Docs: Realtime | Non-realtime • 💰 Pricing: https://int.alibabacloud.com/m/1000411318/ Official blog for details → https://int.alibabacloud.com/m/1000411332/Learn more
