← 全部模型 Claude · 近 7 天Claude Sonnet 50–100 分，越高代表社区使用体验越正面，不代表能力测试成绩。更新于 2026-09-05 16:01 UTC · 2026-08-29 – 2026-09-05 UTC社区口碑分53.1/10021 条评论 · 近 7 天 社区评论精选近 7 天的不同意见，摘录条数不代表真实好评比例。全部摘录正面负面1 条精选摘录 正面编程2026-09-04 “Sonnet 5 真的很好，很有能力。” Redditen机器翻译 展开原文Sonnet 5 is really good and capable.正面通用文本2026-09-04 “Sonnet 5 相当不错” Hacker Newsen机器翻译 展开原文Sonnet 5 is pretty good正面速度与延迟2026-09-03 “Sonnet 5 当时运行正常，顺利完成了全部工作” Redditen机器翻译 展开原文Sonnet 5 was working and got the whole job done just fine负面速度与延迟2026-09-03 “自动模式依赖 sonnet 5 运行，所以我一直在遇到错误” Redditen机器翻译 展开原文the auto mode runs on sonnet 5 and I am getting errors because of that正面通用文本2026-09-02 “Sonnet 5 太棒了” Redditen机器翻译 展开原文Sonnet 5 is awesome负面速度与延迟2026-09-02 “我觉得它比 sonnet 5 更好，尤其是比较速度的时候” Hacker Newsen机器翻译 展开原文i think it's better than sonnet 5, especially when you compare speeds负面通用文本2026-09-02 “sonnet 5 几乎在所有指标上都很差。Anthropic 真的需要一款产品来覆盖低端市场，否则就要被甩在后面了。Sonnet 5 太烂了” Hacker Newsen机器翻译 展开原文sonnet 5 is bad by almost any metric. anthropic really needs something to address the cheaper end of the market before they get left behind. Sonnet 5 sucks正面通用文本2026-09-01 “让另一个模型来解读它创建的任何文档、逻辑或应用，以人类可读的形式呈现，无论是 Sonnet 5 还是 Sol” Redditen机器翻译 展开原文having another model interpret any documentation or logic or apps it creates in something human readable, whether it's Sonnet 5 or Sol负面推理2026-08-31 “我用过的大多数模型在这类问题中都会提到过时的前沿模型。即使是claude sonnet 5也会这样” Redditen机器翻译 展开原文Most models I've used will mention outdated frontier models in these types of questions. Even claude sonnet 5 does this 近 7 天暂时没有符合此筛选条件的精选评论。查看更多摘录
正面速度与延迟2026-09-03 “Sonnet 5 当时运行正常，顺利完成了全部工作” Redditen机器翻译 展开原文Sonnet 5 was working and got the whole job done just fine
负面速度与延迟2026-09-03 “自动模式依赖 sonnet 5 运行，所以我一直在遇到错误” Redditen机器翻译 展开原文the auto mode runs on sonnet 5 and I am getting errors because of that
负面速度与延迟2026-09-02 “我觉得它比 sonnet 5 更好，尤其是比较速度的时候” Hacker Newsen机器翻译 展开原文i think it's better than sonnet 5, especially when you compare speeds
负面通用文本2026-09-02 “sonnet 5 几乎在所有指标上都很差。Anthropic 真的需要一款产品来覆盖低端市场，否则就要被甩在后面了。Sonnet 5 太烂了” Hacker Newsen机器翻译 展开原文sonnet 5 is bad by almost any metric. anthropic really needs something to address the cheaper end of the market before they get left behind. Sonnet 5 sucks
正面通用文本2026-09-01 “让另一个模型来解读它创建的任何文档、逻辑或应用，以人类可读的形式呈现，无论是 Sonnet 5 还是 Sol” Redditen机器翻译 展开原文having another model interpret any documentation or logic or apps it creates in something human readable, whether it's Sonnet 5 or Sol
负面推理2026-08-31 “我用过的大多数模型在这类问题中都会提到过时的前沿模型。即使是claude sonnet 5也会这样” Redditen机器翻译 展开原文Most models I've used will mention outdated frontier models in these types of questions. Even claude sonnet 5 does this
社区评论
精选近 7 天的不同意见，摘录条数不代表真实好评比例。
1 条精选摘录
展开原文
Sonnet 5 is really good and capable.
展开原文
Sonnet 5 is pretty good
展开原文
Sonnet 5 was working and got the whole job done just fine
展开原文
the auto mode runs on sonnet 5 and I am getting errors because of that
展开原文
Sonnet 5 is awesome
展开原文
i think it's better than sonnet 5, especially when you compare speeds
展开原文
sonnet 5 is bad by almost any metric. anthropic really needs something to address the cheaper end of the market before they get left behind. Sonnet 5 sucks
展开原文
having another model interpret any documentation or logic or apps it creates in something human readable, whether it's Sonnet 5 or Sol
展开原文
Most models I've used will mention outdated frontier models in these types of questions. Even claude sonnet 5 does this
近 7 天暂时没有符合此筛选条件的精选评论。