Alibaba Qwen Launches Qwen3.8-LiveTranslate Model Today, Reduces Translation Latency to 2.3 Seconds
According to Alibaba Qwen, the company today launched Qwen3.8-LiveTranslate, a simultaneous interpretation model that reconstructs real-time translation using an Interleave architecture. The model reduces character-level average latency (LAAL) from 2.8 seconds to 2.3 seconds while improving
BABA+3.60%
GateNews·09-19 07:09
