On December 25, according to Gelonghui, Alibaba Cloud Tongyi Qianwen released the industry's first open-source multimodal reasoning model QVQ-72B-Preview. QVQ demonstrates unexpectedly strong visual understanding and reasoning abilities, particularly excelling in solving complex reasoning problems in mathematics, physics, and science. Multiple evaluation metrics show that QVQ surpasses the previous visual understanding model 'open-source champion' Qwen2-VL, with overall performance comparable to reasoning models like 'full version' OpenAIo1 and Claude3.5Sonnet.
阿里云通义开源首个多模态推理模型QVQ
Alibaba Cloud Tongyi has launched the first open-source multimodal reasoning model QVQ.
The translation is provided by third-party software.
The above content is for informational or educational purposes only and does not constitute any investment advice related to Futu. Although we strive to ensure the truthfulness, accuracy, and originality of all such content, we cannot guarantee it.
Hot News
Updated