Bunn (@BunnHack) 在
ai翻译排行榜 中发帖
[i-tested-the-top-models-used-for-translation-on-openrouter-v0-279whdz9j8ue1]
测试了 OpenRouter 上列出的用于翻译的顶级模型,测试了 200 对中英句子。我要求每个模型将一段中文翻译成英文。然后,我用 Comet 对翻译结果进行了排名。令人惊讶的是,LLAMA 3.3 的得分高于 LLAMA 4 Scout,而 LLAMA 3.3 的参数比 Scout 少得多。