阿里升级旗舰版Qwen3模型，性能超过Kimi-K2等领先闭源模型
据通义千问Qwen：今天，我们更新了旗舰版Qwen3模型，推出Qwen3-235B-A22B-FP8非思考模式（Non-thinking）的更新版本，命名为Qwen3-235B-A22B-Instruct-2507-FP8。
新的Qwen3模型，通用能力显著提升，包括指令遵循、逻辑推理、文本理解、数学、科学、编程及工具使用等方面，在GQPA（知识）、AIME25（数学）、LiveCodeBench（编程）、Arena-Hard（人类偏好对齐）、BFCL（Agent能力）等众多测评中表现出色，超过Kimi-K2、DeepSeek-V3等顶级开源模型以及Claude-Opus4-Non-thinking等领先闭源模型。
此外，本次更新的Qwen3模型，还增强了以下关键性能：
-在多语言的长尾知识覆盖方面，模型取得显著进步。
-在主观及开放性任务中，模型显著增强了对用户偏好的契合能力，能够提供更有用的回复，生成更高质量的文本。
-长文本提升到256K，上下文理解能力进一步增强。
目前，Qwen3新模型已在魔搭社区和HuggingFace上开源更新，欢迎开发者们体验。
Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.
Most Discussed
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10