Qwen 2.5: Alibaba's open family spans 0.5B to 72B with Coder and Math variants
In one sentence Alibaba releases Qwen 2.5: 7 sizes (0.5B–72B), updated tokenizer, specialized Coder and Math variants, positioning the family as the open multilingual and code-strong reference.
Alibaba ships a free update of its Qwen model family. Instead of releasing one or two models, seven sizes drop: from a tiny 0.5 billion (runs on a Raspberry Pi) up to a large 72 billion.
There are also specialized versions: Qwen 2.5-Coder for code, Qwen 2.5-Math for math. The 32B Coder beats GPT-4o on code benchmarks.
Weights are downloadable under a license allowing commercial use (except the 72B which requires a contract beyond certain scales). It quickly becomes one of the most used open models worldwide.
Companies
Alibaba
Tools
Qwen2.5, Qwen2.5-Coder, Qwen2.5-Math
Tags
Sources