<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title><![CDATA[Llama 英文强，中文产品能不能主用？]]></title><description><![CDATA[<p dir="auto">公司已经有 Llama 部署经验，但新产品面向中文用户。能不能继续主用 Llama，不换 Qwen 之类？</p>
]]></description><link>https://localaihub.com/topic/95/llama-英文强-中文产品能不能主用</link><generator>RSS for Node</generator><lastBuildDate>Wed, 03 Jun 2026 17:50:13 GMT</lastBuildDate><atom:link href="https://localaihub.com/topic/95.rss" rel="self" type="application/rss+xml"/><pubDate>Wed, 06 May 2026 00:41:00 GMT</pubDate><ttl>60</ttl><item><title><![CDATA[Reply to Llama 英文强，中文产品能不能主用？ on Wed, 06 May 2026 21:34:00 GMT]]></title><description><![CDATA[<p dir="auto">还有客户别看到中英混杂。这个要单独测。</p>
]]></description><link>https://localaihub.com/post/708</link><guid isPermaLink="true">https://localaihub.com/post/708</guid><dc:creator><![CDATA[melo]]></dc:creator><pubDate>Wed, 06 May 2026 21:34:00 GMT</pubDate></item><item><title><![CDATA[Reply to Llama 英文强，中文产品能不能主用？ on Wed, 06 May 2026 18:33:00 GMT]]></title><description><![CDATA[<p dir="auto">记得让真实客服同事盲评，他们最能看出“像不像人话”。</p>
]]></description><link>https://localaihub.com/post/707</link><guid isPermaLink="true">https://localaihub.com/post/707</guid><dc:creator><![CDATA[小高]]></dc:creator><pubDate>Wed, 06 May 2026 18:33:00 GMT</pubDate></item><item><title><![CDATA[Reply to Llama 英文强，中文产品能不能主用？ on Wed, 06 May 2026 16:13:00 GMT]]></title><description><![CDATA[<p dir="auto">合理。已有基础设施是优势，但用户体验是验收。</p>
]]></description><link>https://localaihub.com/post/706</link><guid isPermaLink="true">https://localaihub.com/post/706</guid><dc:creator><![CDATA[JasonL]]></dc:creator><pubDate>Wed, 06 May 2026 16:13:00 GMT</pubDate></item><item><title><![CDATA[Reply to Llama 英文强，中文产品能不能主用？ on Wed, 06 May 2026 15:55:00 GMT]]></title><description><![CDATA[<p dir="auto">我们先做中文用户评测。如果 Llama 过不了口吻和术语，再引入 Qwen 做中文主路由。</p>
]]></description><link>https://localaihub.com/post/705</link><guid isPermaLink="true">https://localaihub.com/post/705</guid><dc:creator><![CDATA[abc_1024]]></dc:creator><pubDate>Wed, 06 May 2026 15:55:00 GMT</pubDate></item><item><title><![CDATA[Reply to Llama 英文强，中文产品能不能主用？ on Wed, 06 May 2026 15:46:00 GMT]]></title><description><![CDATA[<p dir="auto">会，所以要把业务口径放在知识和规则层，不要寄托在模型记忆。路由后也要统一输出检查。</p>
]]></description><link>https://localaihub.com/post/704</link><guid isPermaLink="true">https://localaihub.com/post/704</guid><dc:creator><![CDATA[zeroOne]]></dc:creator><pubDate>Wed, 06 May 2026 15:46:00 GMT</pubDate></item><item><title><![CDATA[Reply to Llama 英文强，中文产品能不能主用？ on Wed, 06 May 2026 14:25:00 GMT]]></title><description><![CDATA[<p dir="auto">多模型会不会增加一致性问题？同一个问题不同模型回答口径不同。</p>
]]></description><link>https://localaihub.com/post/703</link><guid isPermaLink="true">https://localaihub.com/post/703</guid><dc:creator><![CDATA[小潘同学]]></dc:creator><pubDate>Wed, 06 May 2026 14:25:00 GMT</pubDate></item><item><title><![CDATA[Reply to Llama 英文强，中文产品能不能主用？ on Wed, 06 May 2026 13:58:00 GMT]]></title><description><![CDATA[<p dir="auto">如果产品里英文资料很多，最终回答中文，可以让模型先读英文再中文解释。这个场景 Llama 未必差。</p>
]]></description><link>https://localaihub.com/post/702</link><guid isPermaLink="true">https://localaihub.com/post/702</guid><dc:creator><![CDATA[nora]]></dc:creator><pubDate>Wed, 06 May 2026 13:58:00 GMT</pubDate></item><item><title><![CDATA[Reply to Llama 英文强，中文产品能不能主用？ on Wed, 06 May 2026 11:58:00 GMT]]></title><description><![CDATA[<p dir="auto">还有 token 效率。中文 token 数更高时，同样上下文窗口装的信息会少，账单也变。</p>
]]></description><link>https://localaihub.com/post/701</link><guid isPermaLink="true">https://localaihub.com/post/701</guid><dc:creator><![CDATA[陈一]]></dc:creator><pubDate>Wed, 06 May 2026 11:58:00 GMT</pubDate></item><item><title><![CDATA[Reply to Llama 英文强，中文产品能不能主用？ on Wed, 06 May 2026 09:27:00 GMT]]></title><description><![CDATA[<p dir="auto">但评测要覆盖真实中文输入，包括错字、口语、半句、地区表达。不要拿标准普通话样本糊弄自己。</p>
]]></description><link>https://localaihub.com/post/700</link><guid isPermaLink="true">https://localaihub.com/post/700</guid><dc:creator><![CDATA[leaf_1997]]></dc:creator><pubDate>Wed, 06 May 2026 09:27:00 GMT</pubDate></item><item><title><![CDATA[Reply to Llama 英文强，中文产品能不能主用？ on Wed, 06 May 2026 08:26:00 GMT]]></title><description><![CDATA[<p dir="auto">对，工程成本也算成本。若 Llama 中文评测能过，没必要为了潮流换。</p>
]]></description><link>https://localaihub.com/post/699</link><guid isPermaLink="true">https://localaihub.com/post/699</guid><dc:creator><![CDATA[林小北]]></dc:creator><pubDate>Wed, 06 May 2026 08:26:00 GMT</pubDate></item><item><title><![CDATA[Reply to Llama 英文强，中文产品能不能主用？ on Wed, 06 May 2026 07:15:00 GMT]]></title><description><![CDATA[<p dir="auto">维护一套 Llama 比多模型省事，这也是成本。</p>
]]></description><link>https://localaihub.com/post/698</link><guid isPermaLink="true">https://localaihub.com/post/698</guid><dc:creator><![CDATA[小吴]]></dc:creator><pubDate>Wed, 06 May 2026 07:15:00 GMT</pubDate></item><item><title><![CDATA[Reply to Llama 英文强，中文产品能不能主用？ on Wed, 06 May 2026 04:48:00 GMT]]></title><description><![CDATA[<p dir="auto">我们有个双语知识库，英文源码文档 Llama 好，中文客服最后还是 Qwen 更自然。</p>
]]></description><link>https://localaihub.com/post/697</link><guid isPermaLink="true">https://localaihub.com/post/697</guid><dc:creator><![CDATA[melo]]></dc:creator><pubDate>Wed, 06 May 2026 04:48:00 GMT</pubDate></item><item><title><![CDATA[Reply to Llama 英文强，中文产品能不能主用？ on Wed, 06 May 2026 03:18:00 GMT]]></title><description><![CDATA[<p dir="auto">Llama 处理英文技术文档很舒服，中文短答也能做。问题是复杂中文投诉、财务术语、政策解释要测。</p>
]]></description><link>https://localaihub.com/post/696</link><guid isPermaLink="true">https://localaihub.com/post/696</guid><dc:creator><![CDATA[小高]]></dc:creator><pubDate>Wed, 06 May 2026 03:18:00 GMT</pubDate></item><item><title><![CDATA[Reply to Llama 英文强，中文产品能不能主用？ on Wed, 06 May 2026 01:56:00 GMT]]></title><description><![CDATA[<p dir="auto">可以试，但别因为已有部署就忽略中文体验。中文用户对语气和术语很敏感。</p>
]]></description><link>https://localaihub.com/post/695</link><guid isPermaLink="true">https://localaihub.com/post/695</guid><dc:creator><![CDATA[JasonL]]></dc:creator><pubDate>Wed, 06 May 2026 01:56:00 GMT</pubDate></item></channel></rss>