Alibaba Qianwen Full-Modal Large Model Qwen3.5-Omni Launches

BlockBeatNews · 2026-03-30T14:07:23+00:00

Alibaba Qianwen launched the multimodal large model Qwen3.5-Omni on March 30, supporting 256k long context and various audio and video inputs, enhancing multilingual capabilities, able to recognize 113 languages and dialects, and generate 36 types of speech.

BlockBeatNews

2026-03-30 14:07:23

Abstract generation in progress

BlockBeats message, March 30, Alibaba Qianwen announced the launch of its full-modal model Qwen3.5-Omni. The Qwen3.5-Omni series includes Plus, Flash, and Light Instruct versions in three sizes, and supports a 256k long context window. The model supports more than 10 hours of audio input and more than 400 seconds of 720P (1FPS) audio-visual input. The model has undergone native multimodal pretraining on massive text, visual, and audio-visual data totaling over 100 million hours, demonstrating outstanding full-modal perception and generation capabilities. Compared with Qwen3-Omni, Qwen3.5-Omni’s multilingual capabilities have been greatly enhanced, enabling speech recognition in 113 languages and dialects, and speech generation in 36 languages and dialects. (Jin10)

View Original

This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.