The newest Qwen Omni model is Qwen3.8-Omni-Flash, released on September 18, 2026. It has been 50 days since then. The line has only 1 gap between releases so far (172 days), too few to know its usual rhythm.
How long it has been since the last release against the line's usual gap and its recent ones — the rhythm, not a date.
Latest: Qwen3.8-Omni-Flash · Sep 18, 2026
Every Monday, only when something shipped: the AI models that came out, what each one changed, and which lines are due next.
We email you a link to confirm first · unsubscribe anytime · read past issues · about the newsletter · RSS for the Qwen Omni line
Newest first — what each version changed versus the one before, in Alibaba's own numbers.
Qwen's new omni-modal model for agent work: it takes text, images, audio and video, replies in text, and has a 1M-token context window, up from 256K on Qwen3.5-Omni.
Compared with Qwen3.5-Omni-Plus · comparison chosen by Alibaba
More than doubles the earlier Plus model's score on multimodal agent tasks
WildClawBench-MM — real agent tasks involving images, video or audio
71.0
▲ up from 34.5 for Qwen3.5-Omni-Plus · +36.5 points
Costs about 60% less for input and almost 80% less for output than the previous Flash model
Price per 1M tokens — what developers pay, input / output, international pricing
$0.15 / $0.47
▲ down from $0.40 / $2.20 for Qwen3.5-Omni-Flash · text output 79% cheaper
Alibaba's first closed-weight Qwen model: a native omni-modal API model that takes text, images, audio and video, with a 256K-token context window.