{"id":22968,"date":"2024-11-12T19:20:16","date_gmt":"2024-11-12T11:20:16","guid":{"rendered":"https:\/\/www.1ai.net\/?p=22968"},"modified":"2024-11-12T19:20:16","modified_gmt":"2024-11-12T11:20:16","slug":"%e9%98%bf%e9%87%8c%e9%80%9a%e4%b9%89%e5%8d%83%e9%97%ae%e5%bc%80%e6%ba%90-qwen2-5-coder-%e5%85%a8%e7%b3%bb%e5%88%97%e6%a8%a1%e5%9e%8b%ef%bc%8c%e5%8f%b7%e7%a7%b0%e4%bb%a3%e7%a0%81%e8%83%bd%e5%8a%9b","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/22968.html","title":{"rendered":"Ali Tongyi Thousand Questions open source Qwen2.5-Coder full range of models, claiming that the code ability to tie GPT-4o"},"content":{"rendered":"

November 12 news.Ali<\/a>Thousand Questions on Tongyi<\/a>Open Source<\/a> Qwen2.5-Coder full series of models, of which Qwen2.5-Coder-32B-Instruct became the current open source model of SOTA, the official claim that the code ability to tie GPT-4o.<\/p>\n

\"b17c8cdej00smu4rp002bd000v900rsp\"<\/p>\n

Qwen2.5-Coder-32B-Instruct, as the flagship model of this open source, has achieved the best performance among open source models on several popular code generation benchmarks (e.g., EvalPlus, LiveCodeBench, BigCodeBench), and is officially claimed to reach a competitive performance with GPT-4o.<\/p>\n

Ali Tongyi Qianqian previously open source 1.5B, 7B two sizes, this open source brings 0.5B, 3B, 14B, 32B four sizes, covering the mainstream six model sizes.<\/p>\n

The Qwen2.5-Coder 0.5B \/ 1.5B \/ 7B \/ 14B \/ 32B models were used\u00a0Apache 2.0<\/strong>\u00a0license, the 3B model is licensed under the Research Only license.<\/p>\n

\"8cc5598ej00smu4sk0039d000v900iyp\"<\/p>\n

The relevant links are as follows:<\/p>\n