Alibaba's Qwen 1.5 AI Outperforms Major Models
Alibaba's latest AI model Qwen 1.5 outperforms Claude and GPT-3.5 on multiple AI benchmark tests.
Summary
- Alibaba has announced its latest Qwen 1.5 AI model.
- Benchmarks shared on social media platform X show Qwen 1.5 beating Claude and GPT-3.5 on tests like MT-bench and Alapaca-Eval.
- MT-bench tests an AI's ability to answer predefined questions and hold conversations. Qwen 1.5 ranked 4th highest behind only GPT-4 variants.
- On Alapaca-Eval which emulates human interactions, Qwen 1.5 ranked behind only GPT-4 Turbo and HuggingFace's Yi-34B.
- Qwen 1.5 is one of the largest open source AI models with backing from Alibaba's computing resources.
- Many tech giants like Meta, Microsoft, AMD, TSMC etc. are investing heavily in AI given the increased focus in 2024.