GEO

LLMs.txt生成器API弃用指南:从网站内容生成LLM训练文件的工具迁移路径

2026/1/24
LLMs.txt生成器API弃用指南:从网站内容生成LLM训练文件的工具迁移路径
AI Summary (BLUF)

This API generates consolidated text files from websites specifically for LLM training and inference. The service is powered by Firecrawl but will be deprecated after June 30, 2025 in favor of main endpoints. (此API可从网站生成整合文本文件,专为LLM训练和推理设计。该服务由Firecrawl提供支持,但将于2025年6月30日后弃用,建议使用主要端点替代。)

Introduction

The LLMs.txt Generator v2 API has served as a tool for developers to create consolidated text files from website content, specifically formatted for Large Language Model (LLM) training and inference. Powered by Firecrawl, it streamlined the process of preparing web data for AI applications. This post outlines the current status of this API and provides clear guidance for users moving forward.

LLMs.txt Generator v2 API 曾是一个帮助开发者从网站内容生成统一文本文件的工具,这些文件专门为大型语言模型(LLM)的训练和推理而格式化。该工具由 Firecrawl 驱动,简化了为 AI 应用准备网络数据的过程。本文将说明该 API 的当前状态,并为用户提供清晰的后续迁移指南。

API Deprecation Notice

Important Update: This specific API endpoint is now deprecated. The development and maintenance focus has shifted to our main, more robust endpoints. The deprecated endpoint will remain accessible but will not receive updates, bug fixes, or feature enhancements after June 30, 2025.

重要更新:此特定 API 端点现已弃用。开发和维护的重点已转移到我们更强大、更稳定的主端点。被弃用的端点将保持可访问状态,但在 2025 年 6 月 30 日之后 将不再接收更新、错误修复或功能增强。

Key Concepts and Migration Path

The Purpose of LLMs.txt Files

LLMs.txt files are consolidated text extracts from websites, designed to be clean, structured data sources for feeding into LLMs. This facilitates tasks like:

  • Fine-tuning: Training models on specific domain knowledge. (使用特定领域知识训练模型。)
  • Retrieval-Augmented Generation (RAG): Providing context for model inferences. (为模型推理提供上下文。)
  • Content Analysis: Processing web information at scale. (大规模处理网络信息。)

Official Migration Tool

We recommend migrating to the officially supported method for generating LLMs.txt files. The recommended path is to use the dedicated example repository:

Primary Resource: mendableai/create-llmstxt-py on GitHub

This repository contains the canonical, maintained code for generating LLMs.txt files and is aligned with our primary API endpoints.

我们建议迁移到官方支持的生成 LLMs.txt 文件的方法。推荐的路径是使用专门的示例代码库:
主要资源GitHub 上的 mendableai/create-llmstxt-py
该代码库包含生成 LLMs.txt 文件的规范且持续维护的代码,并与我们的主 API 端点保持一致。

Using the Service

To generate a file, you must provide a target URL. For optimal performance and reliability, especially in production environments, using a Firecrawl API key is strongly advised. The service can be accessed via its web interface or integrated programmatically through its API.

要生成文件,您必须提供目标 URL。为了获得最佳性能和可靠性,尤其是在生产环境中,强烈建议使用 Firecrawl API 密钥。该服务可通过其 Web 界面访问,或通过其 API 以编程方式集成。

Main Analysis: Looking Forward

While the deprecated API endpoint was a convenient starting point, the shift to the main endpoints and the GitHub repository represents a strategic move towards greater stability, scalability, and feature parity within our broader developer ecosystem. The new approach offers:

  1. Better Maintenance: Active development and support. (积极的开发和支持。)
  2. Enhanced Features: Access to the latest Firecrawl capabilities. (访问最新的 Firecrawl 功能。)
  3. Community & Transparency: Open-source code allows for inspection and contribution. (开源代码便于检查和贡献。)

Developers currently using the deprecated endpoint should plan their migration to the new repository before the June 2025 end-of-life date to ensure uninterrupted service and access to improvements.

虽然被弃用的 API 端点是一个便捷的起点,但转向主端点和 GitHub 代码库代表了一项战略举措,旨在在我们更广泛的开发者生态系统中实现更高的稳定性、可扩展性和功能一致性。新方法提供了:

  1. 更好的维护:积极的开发和支持。
  2. 增强的功能:访问最新的 Firecrawl 功能。
  3. 社区与透明度:开源代码便于检查和贡献。
    当前使用已弃用端点的开发者应在 2025 年 6 月终止日期之前规划迁移到新代码库,以确保服务不间断并能获取改进。
← 返回文章列表
分享到:微博

版权与免责声明:本文仅用于信息分享与交流,不构成任何形式的法律、投资、医疗或其他专业建议,也不构成对任何结果的承诺或保证。

文中提及的商标、品牌、Logo、产品名称及相关图片/素材,其权利归各自合法权利人所有。本站内容可能基于公开资料整理,亦可能使用 AI 辅助生成或润色;我们尽力确保准确与合规,但不保证完整性、时效性与适用性,请读者自行甄别并以官方信息为准。

若本文内容或素材涉嫌侵权、隐私不当或存在错误,请相关权利人/当事人联系本站,我们将及时核实并采取删除、修正或下架等处理措施。 也请勿在评论或联系信息中提交身份证号、手机号、住址等个人敏感信息。