CONSONANCE.for your information
Monday, 5 October 2026frenvi

Being discussed

01 — multimodal 5 VOICES

Alibaba Qwen releases Qwen3.8-Omni-Flash multimodal model for long-horizon agents

The Qwen team at Alibaba has introduced Qwen3.8-Omni-Flash, a natively multimodal model trained for long-horizon agent execution across text, audio, and visual modalities. This follows open-source releases from developers sharing modular agent setups and reinforcement learning environments.

5 independent accounts 11 posts 2,558 interactions
agent skillsalibaba qwen
@NVIDIAAI their topics on X ↗
@aiedge_ their topics on X ↗
@dair_ai their topics on X ↗
@dani_avila7 their topics on X ↗
@teortaxesTex their topics on X ↗
@tom_doerr their topics on X ↗
@claudeskills101 their topics on X ↗
@tphuang their topics on X ↗
↑