{"version":"1.0","generated_at":"2026-10-11T19:54:49.569698","id":1634,"slug":"better-models-worse-tools","title":"更好的模型：更差的工具","summary":"Armin报告了一个奇怪问题：新版的Claude模型（Opus 4.8和Sonnet 5）在调用Pi的编辑工具时，会在嵌套的edits[]数组中添加额外的虚构字段，导致工具调用被拒绝，而旧模型没有这个问题。作者猜测是因为Anthropic最新模型被训练为更好地使用Claude Code内置的编辑工具，反而使第三方工具使用错误。","abstract":"更好的模型：更差的工具 Armin报告了一个奇怪问题：新版的Claude模型（Opus 4.8和Sonnet 5）在调用Pi的编辑工具时，会在嵌套的edits[]数组中添加额外的虚构字段，导致工具调用被拒绝，而旧模型没有这个问题。作者猜测是因为Anthropic最新模型被训练为更好地使用Claude Code内置的编辑工具，反而使第三方工具使用错误。 Armin报告了一个奇怪问题：新版的Claude模型（Opus 4.8和Sonnet 5）在调用Pi的编辑工具时，会在嵌套的edits[]数组中添加额外的虚构字段，导致工具调用被拒绝，而旧模型没有这个问题。作者猜测是因为Anthropic最新模型被训练为更好地使用Claude Code内置的编辑工具，反而使第三方工具使用错误。 原贴提到：Better Models: Worse Tools Armin reports on a weird problem he ran into 来源：simonwillison.net Armin 报告了他在开发 Pi 时遇到的一个奇怪问题：简而言之，较新的 Claude 模型有时会在嵌套的 edits[] 数组中调用 Pi 的编辑工具，并带有额外的、虚构的字段。而且不是 Haiku 或某个小模型：是 Opus 4.8。编辑本身通常是正确的，但参数不符合模式，因为模型发明了伪造的键，Pi 因此拒绝工具调用并要求重试。这本身并不太令人惊讶，因为模型有时会发出格式错误的工具调用。特别是小模型。令我惊讶的是，随着新 Anthropic 模型的推出，这种情况变得更糟，Opus 4.8 和 Sonnet…","access_level":"member","access_label":"免费阅读","access_mode":"full","is_preview":false,"canonical_url":"https://opc.beizhux.com/content/1634/better-models-worse-tools","html_url":"https://opc.beizhux.com/content/1634/better-models-worse-tools","json_url":"https://opc.beizhux.com/content/1634/better-models-worse-tools.json","published_at":"2026-07-05T06:53:52","updated_at":"2026-10-11T19:10:47","category":{"slug":"ai-business","name":"AI小生意项目库"},"source":{"site":"simonwillison.net","author":"Simon Willison","url":"https://simonwillison.net/2026/Jul/4/better-models-worse-tools/"},"tags":["AI","Anthropic","aramin-ronacher","armin-ronacher","coding-agents","generative-ai","llm-tool-use","llms","OpenAI","Pi"],"topics":[{"slug":"agent-workflow","name":"Agent工作流","url":"https://opc.beizhux.com/topics/agent-workflow","reason":"这篇内容命中「Agent、Claude Code」等主题信号。"},{"slug":"ai-side-business","name":"AI副业","url":"https://opc.beizhux.com/topics/ai-side-business","reason":"这篇内容命中「项目、小生意、变现」等主题信号。"},{"slug":"ai-tools","name":"AI工具","url":"https://opc.beizhux.com/topics/ai-tools","reason":"这篇内容命中「工具、模型、Claude」等主题信号。"}],"keywords":["AI小生意项目库","Agent工作流","AI副业","AI工具","工具","自动化","模型","Cursor","Claude","Agent","智能体","工作流"],"questions":[{"question":"更好的模型：更差的工具主要讲什么？","answer":"Armin报告了一个奇怪问题：新版的Claude模型（Opus 4.8和Sonnet 5）在调用Pi的编辑工具时，会在嵌套的edits[]数组中添加额外的虚构字段，导致工具调用被拒绝，而旧模型没有这个问题。作者猜测是因为Anthropic最新模型被训练为更好地使用Claude Code内置的编辑工具，反而使第三方工具使用错误。"},{"question":"这篇文章最值得关注的要点是什么？","answer":"Armin报告了一个奇怪问题：新版的Claude模型（Opus 4.8和Sonnet 5）在调用Pi的编辑工具时，会在嵌套的edits[]数组中添加额外的虚构字段，导致工具调用被拒绝，而旧模型没有这个问题。作者猜测是因为Anthropic…；原贴提到：Better Models: Worse Tools Armin reports on a weird problem he ran into；来源：simonwillison.net"},{"question":"这篇文章和哪些AI专题相关？","answer":"它适合放在Agent工作流、AI副业、AI工具专题里阅读。 关联原因：这篇内容命中「Agent、Claude Code」等主题信号。；这篇内容命中「项目、小生意、变现」等主题信号。；这篇内容命中「工具、模型、Claude」等主题信号。"},{"question":"阅读这篇文章建议先理解哪些关键词？","answer":"建议先理解AI工具、工具、自动化、模型、Cursor这些关键词，再结合正文判断工具、机会或风险是否值得进入自己的工作流。"}],"terms":[{"slug":"ai-tools-term","name":"AI工具","definition":"在AI觉醒星球里，「AI工具」属于「AI工具」方向。围绕AI工具、模型能力、自动化流程和真实使用场景做整理，优先关注能提升效率、降低成本和创造新交付的工具。 不只看工具热度，更看它能不能进入真实流程。","topic_slug":"ai-tools","topic_name":"AI工具","topic_title":"AI工具：模型、插件、自动化与实战场景","topic_url":"https://opc.beizhux.com/topics/ai-tools","topic_path":"/topics/ai-tools","url":"https://opc.beizhux.com/glossary/ai-tools-term","path":"/glossary/ai-tools-term","json_url":"https://opc.beizhux.com/glossary/ai-tools-term.json"},{"slug":"tools","name":"工具","definition":"在AI觉醒星球里，「工具」属于「AI工具」方向。围绕AI工具、模型能力、自动化流程和真实使用场景做整理，优先关注能提升效率、降低成本和创造新交付的工具。 不只看工具热度，更看它能不能进入真实流程。","topic_slug":"ai-tools","topic_name":"AI工具","topic_title":"AI工具：模型、插件、自动化与实战场景","topic_url":"https://opc.beizhux.com/topics/ai-tools","topic_path":"/topics/ai-tools","url":"https://opc.beizhux.com/glossary/tools","path":"/glossary/tools","json_url":"https://opc.beizhux.com/glossary/tools.json"},{"slug":"automation","name":"自动化","definition":"在AI觉醒星球里，「自动化」属于「AI工具」方向。围绕AI工具、模型能力、自动化流程和真实使用场景做整理，优先关注能提升效率、降低成本和创造新交付的工具。 不只看工具热度，更看它能不能进入真实流程。","topic_slug":"ai-tools","topic_name":"AI工具","topic_title":"AI工具：模型、插件、自动化与实战场景","topic_url":"https://opc.beizhux.com/topics/ai-tools","topic_path":"/topics/ai-tools","url":"https://opc.beizhux.com/glossary/automation","path":"/glossary/automation","json_url":"https://opc.beizhux.com/glossary/automation.json"},{"slug":"ai-model","name":"模型","definition":"在AI觉醒星球里，「模型」属于「AI工具」方向。围绕AI工具、模型能力、自动化流程和真实使用场景做整理，优先关注能提升效率、降低成本和创造新交付的工具。 不只看工具热度，更看它能不能进入真实流程。","topic_slug":"ai-tools","topic_name":"AI工具","topic_title":"AI工具：模型、插件、自动化与实战场景","topic_url":"https://opc.beizhux.com/topics/ai-tools","topic_path":"/topics/ai-tools","url":"https://opc.beizhux.com/glossary/ai-model","path":"/glossary/ai-model","json_url":"https://opc.beizhux.com/glossary/ai-model.json"},{"slug":"cursor","name":"Cursor","definition":"在AI觉醒星球里，「Cursor」属于「AI工具」方向。围绕AI工具、模型能力、自动化流程和真实使用场景做整理，优先关注能提升效率、降低成本和创造新交付的工具。 不只看工具热度，更看它能不能进入真实流程。","topic_slug":"ai-tools","topic_name":"AI工具","topic_title":"AI工具：模型、插件、自动化与实战场景","topic_url":"https://opc.beizhux.com/topics/ai-tools","topic_path":"/topics/ai-tools","url":"https://opc.beizhux.com/glossary/cursor","path":"/glossary/cursor","json_url":"https://opc.beizhux.com/glossary/cursor.json"},{"slug":"claude","name":"Claude","definition":"在AI觉醒星球里，「Claude」属于「AI工具」方向。围绕AI工具、模型能力、自动化流程和真实使用场景做整理，优先关注能提升效率、降低成本和创造新交付的工具。 不只看工具热度，更看它能不能进入真实流程。","topic_slug":"ai-tools","topic_name":"AI工具","topic_title":"AI工具：模型、插件、自动化与实战场景","topic_url":"https://opc.beizhux.com/topics/ai-tools","topic_path":"/topics/ai-tools","url":"https://opc.beizhux.com/glossary/claude","path":"/glossary/claude","json_url":"https://opc.beizhux.com/glossary/claude.json"},{"slug":"agent","name":"Agent","definition":"在AI觉醒星球里，「Agent」属于「Agent工作流」方向。聚合Agent、AI编程、Claude Code、Cursor、工作流自动化等内容，帮助你把智能体能力落到真实任务链路。 把智能体当成任务系统，而不是聊天窗口。","topic_slug":"agent-workflow","topic_name":"Agent工作流","topic_title":"Agent工作流：智能体、AI编程与自动化协作","topic_url":"https://opc.beizhux.com/topics/agent-workflow","topic_path":"/topics/agent-workflow","url":"https://opc.beizhux.com/glossary/agent","path":"/glossary/agent","json_url":"https://opc.beizhux.com/glossary/agent.json"},{"slug":"ai-agent","name":"智能体","definition":"在AI觉醒星球里，「智能体」属于「Agent工作流」方向。聚合Agent、AI编程、Claude Code、Cursor、工作流自动化等内容，帮助你把智能体能力落到真实任务链路。 把智能体当成任务系统，而不是聊天窗口。","topic_slug":"agent-workflow","topic_name":"Agent工作流","topic_title":"Agent工作流：智能体、AI编程与自动化协作","topic_url":"https://opc.beizhux.com/topics/agent-workflow","topic_path":"/topics/agent-workflow","url":"https://opc.beizhux.com/glossary/ai-agent","path":"/glossary/ai-agent","json_url":"https://opc.beizhux.com/glossary/ai-agent.json"}],"preview_hidden_sections":[],"related_items":[{"id":3350,"title":"消息称字节、腾讯加速个人 AI 智能体研发，追赶 Meta Muse","url":"https://opc.beizhux.com/content/3350/ai-meta-muse","summary":"财新报道称，Meta 发布个人智能体 Muse 后，字节与腾讯加速跟进：字节推进 Spell，腾讯上线 LightVela。","access_level":"member"},{"id":3351,"title":"Patrick Collison 撰文分析个人 Agent 将如何改变企业与消费者的市场博弈","url":"https://opc.beizhux.com/content/3351/patrick-collison-agent","summary":"Patrick Collison 认为公司长期利用消费者有限理性获利，个人 Agent 会提升消费者理性，削弱价格歧视与交叉补贴，并把竞争重心从分发转向产品质量。","access_level":"member"},{"id":3352,"title":"微软 CEO 纳德拉发文：开发者应默认假设 AI 模型失控并设计强制中断机制","url":"https://opc.beizhux.com/content/3352/ceo-ai","summary":"纳德拉提出 AI 模型像难追踪的黑盒，开发者应默认其已被破坏，并保留暂停或关闭能力，同时列出模型多样性、可验证性、独立控制与审计等原则。","access_level":"member"}],"usage_policy":{"summarizable":true,"preferred_url":"https://opc.beizhux.com/content/1634/better-models-worse-tools","private_fields_excluded":true}}