Alibaba Cloud recently introduced their own LLM service in the Qwen series, including the Qwen-Max, Qwen-Plus, and Qwen-Turbo versions. The Qwen-Max is being touted as comparable to the Llama3.1-405B and GPT-4o, with successful tests in areas like MATH and LiveCodeBenach.
The cost of using Qwen-Max stands at $10 per million tokens for input and $30 per million tokens for output, making it more expensive than GPT-4o.
In addition to the simultaneous release of other models like Tongyi Wanxiang, which generates images and videos from text, Qwen2-VL for image and video processing, and AI Developer for programming assistance.
TLDR: Alibaba Cloud unveils new LLM service in the Qwen series, with Qwen-Max boasting high performance at a costlier price point compared to competitors.
Leave a Comment