Tongyi Qianwen 72B, 1.8B, and Audio models were released, following the example of Meta and Alibaba Cloud to open source the Tongyi Qianwen 72 billion parameter model Qwen-72B, 1

Tongyi Qianwen 72B, 1.8B, and Audio models were released, following Meta's example of flipping the table

Alibaba Cloud has open-sourced the Qwen-72B 72 billion parameter model, the Qwen-1.8 billion parameter model Qwen-1.8B, and the Qwen-Audio large audio model. The three models are Transformer-based large language models trained on ultra-large-scale pre-trained data. Qwen-72B is effective in multiple Chinese and English downstream evaluation tasks, Qwen-1.8B supports low-cost deployment, and Qwen-Audio supports long text processing. Users can download relevant models through the AI Express website.