Alibaba Cloud launches new visual-language model Qwen2.5-VL with advanced capabilities

From Nasdaq: 2025-01-29 12:49:44

Alibaba Cloud introduces Qwen2.5-VL, a new visual-language model with improved performance over its predecessor. The model comes in different parameter sizes and outperforms competitors like GPT-4o and DeepSeek-V3. Qwen2.5-VL-72B-Instruct is now available on Qwen Chat, while the series can be accessed on Hugging Face and Model Scope.

Qwen2.5-VL showcases advanced multimodal capabilities, excelling in visual comprehension of various media types. The model can understand long videos, answer video-related questions, and accurately identify specific segments within videos. Alibaba claims it is a significant advancement in AI technology.



Read more at Nasdaq: Alibaba Cloud Releases Latest AI Models For Enhanced Visual Understanding