← All model drops
Live drop
Qwen3.8-Omni-Flash
Qwen (Alibaba)Dropped Sep 18, 2026
Native omnimodal agent that plans, tools, and delivers finished audio-video workflows from long inputs.
About this release
Live drop
Qwen (Alibaba)Dropped Sep 18, 2026
Native omnimodal agent that plans, tools, and delivers finished audio-video workflows from long inputs.
About this release
Qwen3.8-Omni-Flash is Alibaba's next-generation native omnimodal model accepting text, image, audio, and video inputs with a 1M-token context window. It advances from perception to agentic execution across coding, GUI, and especially audio-video tasks such as video editing, MV production, film commentary, summarization, and real-time conversation. It ships with expanded Qwen-MM-Plugins and open-sourced Qwen-Live Harness for long-horizon workflows.
Read the official Qwen (Alibaba) launch pageFirst builds incoming — this lane fills in automatically as people start building with Qwen3.8-Omni-Flash. Know one? Submit the URL.