发布说明#
n8n 的新功能和错误修复。
您也可以在 GitHub 仓库中查看发布版本。
Latest 和 Next 版本
n8n 每周发布一个新的次要版本。latest 版本用于生产环境。next 是最新的发布版本。您应该将 next 视为测试版：它可能不稳定。要报告问题，请使用论坛。
Current latest: 1.112.4
Current next: 1.113.1
如何更新 n8n#
更新 n8n 的步骤取决于您使用的 n8n 平台。请参阅您的 n8n 相关文档：
n8n 中的语义化版本控制#
n8n 使用语义化版本控制。所有版本号都采用 MAJOR.MINOR.PATCH 格式。版本号递增规则如下：
- MAJOR 版本：进行不兼容的更改时，可能需要用户操作。
- MINOR 版本：以向后兼容的方式添加功能时。
- PATCH 版本：进行向后兼容的错误修复时。
旧版本
您可以在这里找到 n8n 旧版本的发布说明
[email protected]#
查看此版本的提交记录。
发布日期： 2025-09-23
Next 版本
这是 next 版本。n8n 建议使用 latest 版本。next 版本可能不稳定。要报告问题，请使用论坛。
此版本包含错误修复。
数据表#
我们很高兴推出数据表，为 n8n 带来内置数据存储功能。您现在可以直接在平台内存储和查询结构化数据，而无需依赖外部数据库来处理许多常见的自动化场景。跟踪运行之间的工作流状态、存储令牌或会话数据、维护产品或客户参考表，或为多步骤流程暂存中间结果。
以前，持久化数据意味着配置和连接到外部存储（如 Redis 或 Google Sheets）。这增加了凭据设置、基础设施开销、延迟和持续上下文切换。数据表消除了这种摩擦，使您的数据易于编辑并贴近您的工作流。
数据表现已在所有套餐中提供。目前支持数字、字符串和日期时间类型，后续将很快支持 JSON。在云端，每个实例最多可存储 50 MB。自托管部署默认也是 50 MB，但如果您的基础设施允许，可以调整此限制。
使用方法#
创建数据表#
- 从画布中，打开创建工作流下拉菜单并选择创建数据表。
- 或者，转到左侧导航栏的概览面板并打开数据表选项卡。
在工作流中使用数据表#
- 将数据表节点添加到您的工作流中以获取、更新、插入、更新插入或删除行。
调整存储限制（仅限自托管）#
- 使用环境变量更改默认的 50 MB 限制：
N8N_DATA_TABLES_MAX_SIZE_BYTES。查看配置文档。
注意事项#
- 数据表目前不支持外键或默认值。
- 目前，项目中的所有用户都可以访问所有数据表。计划提供更细粒度的权限和共享选项。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-09-23
最新版本
这是 latest 版本。n8n 建议使用 latest 版本。next 版本可能不稳定。要报告问题，请使用论坛。
此版本包含编辑器改进。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-09-22
此版本包含核心更新、编辑器改进、新节点、节点更新和错误修复。
贡献者#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-09-19
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-09-19
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-09-19
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-09-18
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-09-17
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-09-15
此版本包含 API 改进、核心更新、编辑器改进、节点更新和错误修复。
贡献者#
GuraaseesSingh
jabbson
ongdisheng
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-09-08
此版本包含核心更新、API 改进、节点更新和错误修复。
贡献者#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-09-03
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-09-03
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-09-01
此版本包含核心更新、编辑器改进、节点更新、性能优化和错误修复。
贡献者#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-08-27
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-08-27
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-08-25
此版本包含核心更新、编辑器改进、节点更新、性能优化和错误修复。
贡献者#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-08-20
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-08-20
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-08-18
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-08-18
此版本包含新的 CLI 工具、编辑器改进、节点更新、性能优化和错误修复。
工作流差异#
对于在不同环境中工作的团队，部署通常涉及多个人在不同时间进行更改。如果没有清晰的更改视图，很容易遗漏一些重要的内容。
工作流差异为您提供了一种简单且可视化的方式，在环境之间部署工作流更改之前进行审查。
通过它，您可以：
- 通过清晰的颜色突出显示，快速查看添加、更改或删除的内容。
- 轻松查看工作流上的重要设置更改。
- 检查每个节点内部的更改，并通过并排查看其设置来发现连接器更新。
- 快速统计所有更改以了解部署的规模。
工作流差异简化了部署前更改的审查和批准过程，使团队能够在不破坏现有自动化或中断生产的情况下协作处理工作流。这是在 n8n 中集成 DevOps 最佳实践的进一步发展。
现在可供使用环境的企业客户使用。
贡献者#
ManuLasker
EternalDeiwos
jreyesr
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-08-15
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-08-14
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-08-11
此版本包含一个向后移植的更新。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-08-11
此版本包含错误修复。
贡献者#
Amsal1
andrewzolotukhin
DMA902
fkowal
Gulianrdgd
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-08-08
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-08-07
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-08-07
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-08-05
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-08-04
此版本包含性能改进、核心更新、编辑器改进、节点更新、一个新节点和错误修复。
取消活跃工作流限制和全新的自托管商业计划#
我们推出了全新的定价模型，让各种规模的构建者都能更轻松地采用和扩展 n8n 自动化。
新功能
取消活跃工作流限制。
从入门版到企业版，所有 n8n 计划现在都包含无限的用户、工作流和步骤。我们的定价基于执行量。这意味着您可以构建和测试任意数量的工作流，包括复杂的、数据密集型或长时间运行的自动化，而无需担心配额问题。
为成长型团队推出全新的自托管商业计划
专为中小型企业和中型公司设计，商业计划包含以下功能：
- 6 个共享项目
- SSO、SAML 和 LDAP
- 不同的环境
- 全局变量
- 使用 Git 的版本控制
- 30 天的 Insights 洞察
请注意，此计划仅包含来自我们社区论坛的支持。如需专属支持，我们建议升级到企业计划。
企业版定价现在基于执行量扩展
企业计划不再使用基于工作流的定价，现在也基于执行量。
您需要做的事情
为确保这些更改适用于您的账户，请将您的 n8n 实例更新到最新版本。
阅读博客了解完整详情。
贡献者#
baruchiro
killthekitten
baileympearson
Yingrjimsch
joshualipman123
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-08-01
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-08-01
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-07-31
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-07-28
此版本包含核心更新、编辑器改进、节点更新和错误修复。
响应聊天节点#
使用响应聊天节点，您现在可以在n8n聊天中原生访问人机交互功能。
启用对话体验，您可以在其中请求澄清、在采取进一步行动前请求批准，并获得中间结果——所有这些都在单个工作流执行中完成。
这解锁了感觉更自然的多轮交互，并减少了所需的执行次数。它非常适合构建交互式AI用例，如对话表单、基于用户回复的分支工作流和逐步批准。
🛠️ 操作方法：
- 添加一个聊天触发器节点，并为响应模式选择使用响应节点
- 将响应聊天节点放在工作流中的任何位置，向聊天发送消息，并可选择在继续执行工作流步骤之前等待用户输入响应。
贡献者#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-07-23
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-07-22
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-07-21
此版本包含核心更新、编辑器改进、一个新节点、节点更新和错误修复。
贡献者#
nunulk
iaptsiauri
KGuillaume-chaps
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-07-18
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-07-17
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-07-17
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-07-14
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-07-14
此版本包含核心更新、编辑器改进、新节点、节点改进和错误修复。
聊天流式传输#
使用 n8n 聊天界面时不再需要等待完整响应加载。流式传输现在可以逐字传递 AI 生成的文本回复，让用户可以在消息生成时阅读。这感觉更快、更流畅，更符合用户对聊天体验的期望。
流式传输在公共聊天视图（托管或嵌入式）中可用，还可以通过 webhook 在自定义应用中使用。
🛠️ 操作方法
在以下节点的节点详细信息视图中配置流式传输：
- 聊天触发器节点：选项>添加字段>响应模式>流式传输
- Webhook 节点：响应>流式传输
- AI 代理节点：添加选项>启用流式传输
改进的实例用户列表，可见性更高#
实例用户列表已更新，采用新的表格布局和其他详细信息，帮助管理员更轻松地管理访问权限。
您现在可以：
- 查看用户总数并按姓名或电子邮件筛选
- 查看每个用户有权访问的项目
- 查看用户是否启用了 2FA 并基于此排序
- 查看每个用户的最后活跃日期
这使得审计用户活动、识别不活跃账户以及了解访问权限在实例中的分布变得更加容易。
Webhook HTML 响应#
从此版本开始，如果您的工作流向 webhook 发送 HTML 响应，n8n 会自动将内容包装在 <iframe> 中。这是保护实例用户的安全机制。
这会产生以下影响：
- HTML 在沙盒化的 iframe 中渲染，而非直接在父文档中渲染。
- 尝试访问顶级窗口或本地存储的 JavaScript 代码将失败。
- 沙盒化的 iframe 中无法使用身份验证标头（例如基本认证）。您需要使用替代方法，比如在 HTML 中嵌入短期访问令牌。
- 相对 URL（例如
<form action="/">）将不起作用。请改用绝对 URL。
AI 评估内置指标#
使用评估是任何 AI 解决方案的最佳实践，如果可靠性和可预测性对业务至关重要，那么评估是必不可少的。在此版本中，我们通过引入一套内置指标，让在 n8n 中设置评估变得更加容易。这些指标可以审查 AI 响应，并根据正确性、有用性等因素分配分数。
您可以定期运行评估并查看随时间变化的分数，作为监控 AI 工作流性能的方式。您还可以比较不同模型之间的结果以指导模型选择，或在提示更改前后运行评估以支持数据驱动的迭代构建。
与 n8n 中的所有评估一样，您需要一个包含要测试输入的数据集。对于某些评估，数据集还必须包含期望的输出（基准真相）进行比较。评估工作流将每个输入通过您正在测试的部分来生成响应。内置指标根据您正在测量的方面为每个响应评分，允许您比较更改前后的结果或在评估选项卡中跟踪随时间的趋势。
您仍然可以定义自己的自定义指标，但对于常见用例，内置选项使实施更加快速。
🛠️ 操作方法：
- 按照这里的描述设置您的评估，使用Evaluation 评估节点作为触发器，另一个使用Set Metrics 设置指标操作。
- 在Set Metrics 设置指标节点中，从下拉列表中选择一个指标。
- 定义您所选指标所需的任何其他参数。在大多数情况下，这包括将数据集列映射到相应字段。
📏 可用的内置指标：
- Correctness (AI-based) 正确性（基于AI）： 将AI工作流生成的响应与预期答案进行比较。另一个LLM充当评判者，根据您在提示中提供的指导对响应进行评分。
- Helpfulness (AI-based) 有用性（基于AI）： 使用LLM和提示定义的评分标准评估响应相对于用户查询的有用程度。
- String Similarity 字符串相似性： 通过比较字符串来衡量响应与预期输出的匹配程度。适用于命令生成或输出需要遵循特定结构的情况。
- Categorization 分类： 检查响应是否匹配预期标签，例如将项目分配到正确的类别。
- Tools Used 使用的工具： 验证AI代理是否调用了您在数据集中指定的工具。 要启用此功能，请确保在您的代理中开启Return Intermediate Steps 返回中间步骤，以便评估可以访问它实际调用的工具。
🧠 Keep in mind
- Registered Community Edition enables analysis of one evaluation in the Evaluations tab which allows easy comparison of evaluation runs over time. Pro and Enterprise plans allow unlimited evaluations in the Evaluations tab.
Learn more about setting up and customizing evaluations.
AI代理工具节点#
通过AI代理工具节点，我们引入了多代理编排的简化模式，可以在单个执行中运行并完全保持在一个画布上。您现在可以将多个AI代理工具节点连接到主要的AI代理节点，允许它监督和委派工作给其他专业代理。
这种设置特别适用于构建像现实世界团队一样运作的复杂系统，其中主导代理将任务的部分分配给专家。您甚至可以添加多层代理来指导其他代理，就像您在真实的多层组织结构中一样。它还有助于提示管理，让您将长而复杂的指令分解为跨多个代理的较小、专注的任务。虽然使用子工作流已经可以实现类似的编排，但当您希望交互在单个执行中发生或更喜欢从单个画布管理和调试所有内容时，AI代理工具节点是一个很好的选择。
🛠️ 操作方法：
- 向您的工作流添加一个AI Agent AI代理节点，然后点击+创建工具连接。
- 从节点面板中搜索并选择AI Agent Tool AI代理工具节点。
- 清楚地命名节点，以便主要代理可以引用它，然后添加简短描述和提示。
- 连接代理执行其角色所需的任何LLM、内存和工具。
- 指导主要的AI Agent AI代理何时使用AI Agent Tool AI代理工具并在其提示中传递相关上下文。
🧠 请记住：
- 编排代理默认不传递完整的执行上下文。任何必要的上下文都必须包含在提示中。
AI Agent Tool AI代理工具节点使构建分层、代理到代理的工作流变得更容易，而无需依赖子工作流，帮助您在构建和调试多代理系统时更快地移动。
贡献者#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-07-11
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-07-11
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-07-09
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-07-07
此版本包含核心更新、编辑器改进、新节点、节点更新和错误修复。
在您的实例中强制启用双重认证#
企业实例所有者现在可以为其实例中的所有用户强制启用双重认证 (2FA)。
启用后，任何尚未设置 2FA 的用户都将被重定向完成设置，然后才能继续使用 n8n。这有助于组织符合内部安全政策，并确保整个工作区获得更强的保护。
此功能仅在企业版计划中可用。
贡献者#
marty-sullivan
cesars-gh
dudanogueira
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-07-03
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-06-30
此版本包含核心更新、编辑器改进、节点更新和错误修复。
贡献者#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-06-25
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
查看此版本的提交记录。
发布日期： 2025-06-23
此版本包含核心更新、编辑器改进、一个新节点、节点更新和错误修复。
模型选择器节点#
模型选择器节点 让您在工作流中使用多个 LLM 时拥有更多控制权。
使用它来确定哪个连接的模型应该处理给定的输入，基于表达式或全局变量等条件。这使得实施模型路由策略变得更容易，例如根据性能、任务类型、成本或可用性切换模型。
🛠️ 操作方法： 将多个模型节点连接到模型选择器节点，然后在节点设置中配置路由条件。
🧠 请记住：
- 规则按顺序评估。第一个匹配的规则决定使用哪个模型，即使其他规则也会匹配。
- 作为子节点，表达式在这里的行为不同：它们总是解析为第一个项目，而不是依次为每个项目解析。
模型选择器节点在评估或生产场景中特别有用，其中模型之间的路由逻辑需要根据性能、成本、可用性或数据集特定需求进行调整。
支持 OIDC (OpenID Connect) 身份验证#
您现在可以使用 OIDC (OpenID Connect) 作为单点登录 (SSO) 的身份验证方法。
这为企业团队提供了更多灵活性，可以使用广泛采用且易于管理的标准将 n8n 与其现有身份提供者集成。OIDC 现在与 SAML 一起提供，为企业提供选择最适合其内部需求的选项。
项目管理员现在可以在环境中提交到 Git#
项目管理员现在可以通过环境功能直接将工作流和凭据更改提交到 Git。此更新通过为项目级管理员提供对提交更改的直接控制来简化工作流部署过程。它还确保最了解其工作流的人员可以自行审查和提交更新，而无需涉及实例级管理员。
贡献者#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-06-19
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-06-18
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-06-16
此版本包含性能改进、核心更新、编辑器更改、节点更新和错误修复。
自动命名节点#
默认节点名称现在会根据选择的资源和操作自动更新，因此您始终可以一目了然地知道节点的作用。
这为您的画布增加了清晰度，并节省了手动重命名节点的时间。
不用担心，自动命名不会破坏引用。而且，如果您自己重命名了节点，我们会保持您编写的方式。
Support for RAG extended with built-in templates#
Retrieval-Augmented Generation (RAG) can improve AI responses by providing language models access to data sources with up-to-date, domain-specific, or proprietary knowledge. RAG workflows typically rely on vector stores to manage and search this data efficiently.
To get the benefits of using vector stores, such as returning results based on semantic meaning rather than just keyword matches, you need a way to upload your data to the vector store and a way to query it.
In n8n, uploading and querying vectors stores happens in two workflows. Now, you have an example to get your started and make implementation easier with the RAG starter template.
- The Load Data workflow shows how to add data with the appropriate embedding model, split it into chunks with the Default Data Loader, and add metadata as desired.
- The Retriever workflow for querying data, shows how agents and vector stores work together to help you define highly relevant results and save tokens using the Question and Answer tool.
Enable semantic search and the retrieval of unstructured data for increased quality and relevance of AI responses.
🛠️ How to:
- Search for RAG starter template in the search bar of the Nodes panel to insert it into your workflow.
Learn more about implementing RAG in n8n here.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-06-12
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-06-11
This release contains performance improvements, core updates, editor changes, node updates, a new node, and bug fixes.
Contributors#
luka-mimi
Alexandero89
khoazero123
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-06-04
This release contains backports.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-06-03
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-06-02
This release contains new features, performance improvements and bug fixes.
Convert to sub-workflow#
Large, monolithic workflows can slow things down. They’re harder to maintain, tougher to debug, and more difficult to scale. With sub-workflows, you can take a more modular approach, breaking up big workflows into smaller, manageable parts that are easier to reuse, test, understand, and explain.
Until now, creating sub-workflows required copying and pasting nodes manually, setting up a new workflow from scratch, and reconnecting everything by hand. Convert to sub-workflow allows you to simplify this process into a single action, so you can spend more time building and less time restructuring.
How it works
- Highlight the nodes you want to convert to a sub-workflow. These must:
- Be fully connected, meaning no missing steps in between them
- Start from a single starting node
- End with a single node
- Right-click to open the context menu and select Convert to sub-workflow
- Or use the shortcut:
Alt + X
- Or use the shortcut:
- n8n will:
- Open a new tab containing the selected nodes
- Preserve all node parameters as-is
- Replace the selected nodes in the original workflow with a Call My Sub-workflow node
Note: You will need to manually adjust the field types in the Start and Return nodes in the new sub-workflow.
This makes it easier to keep workflows modular, performant, and easier to maintain.
Learn more about sub-workflows.
This release contains performance improvements and bug fixes.
Contributors#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-06-02
Build failure
This release failed to build. Please use 1.97.0 instead.
This release contains API updates, core changes, editor improvements, node updates, and bug fixes.
API support for assigning users to projects#
You can now use the API to add and update users within projects. This includes:
- Assigning existing or pending users to a project with a specific role
- Updating a user’s role within a project
- Removing users from one or more projects
This update now allows you to use the API to add users to both the instance and specific projects, removing the need to manually assign them in the UI.
Add pending users to project member assignment#
You can now add pending users, those who have been invited but haven't completed sign-up, to projects as members.
This change lets you configure a user's project access upfront, without waiting for them to finish setting up their account. It eliminates the back-and-forth of managing access post-sign-up, ensuring users have the right project roles immediately upon joining.
Contributors#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-05-29
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-05-27
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-05-27
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-05-26
This release contains core updates, editor improvements, node updates, and bug fixes.
Evaluations for AI workflows#
We’ve added a feature to help you iterate, test, and compare changes to your AI automations before pushing them to production so you can achieve more predictability and make better decisions.
When you're building with AI, a small prompt tweak or model swap might improve results with some inputs, while quietly degrading performance with others. But without a way to evaluate performance across many inputs, you’re left guessing whether your AI is actually getting better when you make a change.
By implementing Evaluations for AI workflows in n8n, you can assess how your AI performs across a range of inputs by adding a dedicated path in your workflow for running test cases and applying custom metrics to track results. This helps you build viable proof-of-concepts quickly, iterate more effectively, catch regressions early, and make more confident decisions when your AI is in production.
Evaluation node and tab#
The Evaluation node includes several operations that, when used together, enable end-to-end AI evaluation.
Use this node to:
- Run your AI logic against a wide range of test cases in the same execution
- Capture the outputs of those test cases
- Score the results using your own metrics or LLM-as-judge logic
- Isolate a testing path to only include the nodes and logic you want to evaluate
The Evaluations tab enables you to review test results in the n8n UI, perfect for comparing runs, spotting regressions, and viewing performance over time.
🛠 How evaluations work#
The evaluation path runs alongside your normal execution logic and only activates when you want—making it ideal for testing and iteration.
Get started by selecting an AI workflow you want to evaluate that includes one or more LLM or Agent nodes.
-
Add an Evaluation node with the On new Evaluation event operation. This node will act as an additional trigger you’ll run only when testing. Configure it to read your dataset from Google Sheets, with each row representing a test input.
💡 Better datasets mean better evaluations. Craft your dataset from a variety of test cases, including edge cases and typical inputs, to get meaningful feedback on how your AI performs. Learn more and access sample datasets here.
-
Add a second Evaluation node using the Set Outputs operation after the part of the workflow you're testing—typically after an LLM or Agent node. This captures the response and writes it back to your dataset in Google Sheets.
-
To evaluate output quality, add a third Evaluation node with the Set Metrics operation at a point after you’ve generated the outputs. You can develop workflow logic, custom calculations, or add an LLM-as-Judge to score the outputs. Map these metrics to your dataset in the node’s parameters.
💡 Well-defined metrics = smarter decisions. Scoring your outputs based on similarity, correctness, or categorization can help you track whether changes are actually improving performance. Learn more and get links to example templates here.
When the Evaluation trigger node is executed, it runs each input in our dataset through your AI logic. This continues until all test cases are processed, a limit is reached, or you manually stop the execution. Once your evaluation path is set up, you can update your prompt, model, or workflow logic—and re-run the Evaluation trigger node to compare results. If you’ve added metrics, they’ll appear in the Evaluations tab.
In some instances, you may want to isolate your testing path to make iteration faster or to avoid executing downstream logic. In this case, you can add an Evaluation node with the Check If Evaluating operation to ensure only the expected nodes run when performing evaluations.
Things to keep in mind#
Evaluations for AI Workflows are designed to fit into your development flow, with more enhancements on the way. For now, here are a few things to note:
- Test datasets are currently managed through Google Sheets. You’ll need a Google Sheets credential to run evaluations.
- Each workflow supports one evaluation at a time. If you’d like to test multiple segments, consider splitting them into sub-workflows for more flexibility.
- Community Edition supports one single evaluation. Pro and Enterprise plans allow unlimited evaluations.
- AI Evaluations are not enabled for instances in scaling mode at this time.
You can find details, tips, and common troubleshooting info here.
👉 Learn more about the AI evaluation strategies and practical implementation techniques. Watch now.
Contributors#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-05-19
This release contains editor improvements, an API update, node updates, new nodes, and bug fixes.
Verified community nodes on Cloud#
We’ve expanded the n8n ecosystem and unlocked a new level of flexibility for all users including those on n8n Cloud! Now you can access a select set of community nodes and partner integrations without leaving the canvas. This means you install and automate with a wider range of integrations without leaving your workspace. The power of the community is now built-in.
This update focuses on three major improvements:
- Cloud availability: Community nodes are no longer just for self-hosted users. A select set of nodes is now available on n8n Cloud.
- Built-in discovery: You can find and explore these nodes right from the Nodes panel without leaving the editor or searching on npm.
- Trust and verification: Nodes that appear in the editor have been manually vetted for quality and security. These verified nodes are marked with a checkmark.
We’re starting with a selection of around 25 nodes, including some of the most-used community-built packages and partner-supported integrations. For this phase, we focused on nodes that don’t include external package dependencies - helping streamline the review process and ensure a smooth rollout.
This is just the start. We plan to expand the library gradually, bringing even more verified nodes into the editor along with the powerful and creative use cases they unlock. In time, our criteria will evolve, opening the door to a wider range of contributions while keeping quality and security in focus.
Learn more about this update and find out which nodes are already installable from the editor in our blog post.
💻 Use a verified node
Make sure you're on n8n version 1.94.0 or later and the instance Owner has enabled verified community nodes. On Cloud, this can be done from the Admin Panel. For self-hosted instances, please refer to documentation. In both cases, verified nodes are enabled by default.
- Open the Nodes panel from the editor
- Search for the Node. Verified nodes are indicated by a shield 🛡️
- Select the node and click Install
Once an Owner installs a node, everyone on the instance can start using it—just drag, drop, and connect like any other node in your workflow.
🛠️ Build a node and get it verified
Want your node to be verified and discoverable from the editor? Here’s how to get involved:
- Review the community node verification guidelines.
- If you’re building something new, follow the recommendations for creating nodes.
- Check your design against the UX guidelines.
- Submit your node to npm.
- Request verification by filling out this form.
Already built a node? Raise your hand!
If you’ve already published a community node and want it considered for verification, make sure it meets the requirements noted above, then let us know by submitting the interest form. We’re actively curating the next batch and would love to include your work.
Extended logs view#
When workflows get complex, debugging can get... clicky. That’s where an extended Logs View comes in. Now you can get a clearer path to trace executions, troubleshoot issues, and understand the behavior of a complete workflow — without bouncing between node detail views.
This update brings a unified, always-accessible panel to the bottom of the canvas, showing you each step of the execution as it happens. Whether you're working with loops, sub-workflows, or AI agents, you’ll see a structured view of everything that ran, in the order it ran—with input, output, and status info right where you need it.
You can jump into node details when you want to dig deeper, or follow a single item through every step it touched. Real-time highlighting shows you which nodes are currently running or have failed, and you’ll see total execution time for any workflow—plus token usage for AI workflows to help monitor performance. And if you're debugging across multiple screens? Just pop the logs out and drag them wherever you’d like.
⚙️What it does
- Adds a Logs view to the bottom of the canvas that can be opened or collapsed. (Chat also appears here if your workflow uses it).
- Displays a hierarchical list of nodes in the order they were executed—including expanded views of sub-workflows.
- Allows you to click a node in hierarchy to preview inputs and outputs directly, or jump into the full Node Details view with a link.
- Provides ability to toggle input and output data on and off.
- Highlights each node live as it runs, showing when it starts, completes, or fails.
- Includes execution history view to explore past execution data in a similar way.
- Shows roll-up stats like total execution time and total AI tokens used (for AI-enabled workflows).
- Includes a “pop out” button to open the logs as a floating window—perfect for dragging to another screen while debugging.
🛠️How to
To access the expanded logs view, click on the Logs bar at the bottom of the canvas. The view is also opens up when you open the chat window on the bottom of the page.
Contributors#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-05-12
This release contains core updates, editor improvements, new nodes, node updates, and bug fixes.
Faster ways to open sub-workflows#
We’ve added several new ways to navigate your multi-workflow automations faster.
From any workflow with a sub-workflow node:
🖱️ Right-click on a sub-workflow node and select Open sub-workflow from the context menu
⌨️ Keyboard shortcuts
- Windows:
CTRL + SHIFT + OorCTRL + Double Click - Mac:
CMD + SHIFT + OorCMD + Double Click
These options will bring your sub-workflow up in a new tab.
Archive workflows#
If you’ve ever accidentally removed a workflow, you’ll appreciate the new archiving feature. Instead of permanently deleting workflows with the Remove action, workflows are now archived by default. This allows you to recover them if needed.
How to:
- Archive a workflow - Select Archive from the Editor UI menu. It has replaced the Remove action.
- Find archived workflows - Archived workflows are hidden by default. To find your archived workflows, select the option for Show archived workflows in the workflow filter menu.
- Permanently delete a workflow - Once a workflow is archived, you can Delete it from the options menu.
- Recover a workflow - Select Unarchive from the options menu.
Keep in mind:
- Workflows archival requires the same permissions as required previously for removal.
- You cannot select archived workflows as sub-workflows to execute
- Active workflows are deactivated when they are archived
- Archived workflows can not be edited
Contributors#
LeaDevelop
ayhandoslu
valentina98
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-05-08
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-05-08
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-05-06
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-05-05
This release contains core updates, editor improvements, node updates, and bug fixes.
Partial Execution for AI Tools#
We’ve made it easier to build and iterate on AI agents in n8n. You can now run and test specific tools without having to execute the entire agent workflow.
Partial execution is especially useful when refining or troubleshooting parts of your agent logic. It allows you to test changes incrementally, without triggering full agent runs, reducing unnecessary AI calls, token usage, and downstream activity. This makes iteration faster, more cost-efficient, and more precise when working with complex or multi-step AI workflows.
Partial execution for AI tools is available now for all tools - making it even easier to build, test, and fine-tune AI agents in n8n.
How to:
To use this feature you can either:
- Click the Play button on the tool you want to execute directly from the canvas view.
- Open the tool’s Node Details View and select "Execute Step" to run it from there.
If you have previously run the workflow, the input and output will be prefilled with data from the last execution. A pop-up form will open where you can manually fill in the parameters before executing your test.
Extended logs view#
When workflows get complex, debugging can get... clicky. That’s where an extended Logs View comes in. Now you can get a clearer path to trace executions, troubleshoot issues, and understand the behavior of a complete workflow — without bouncing between node detail views.
This update brings a unified, always-accessible panel to the bottom of the canvas, showing you each step of the execution as it happens. Whether you're working with loops, sub-workflows, or AI agents, you’ll see a structured view of everything that ran, in the order it ran—with input, output, and status info right where you need it.
You can jump into node details when you want to dig deeper, or follow a single item through every step it touched. Real-time highlighting shows you which nodes are currently running or have failed, and you’ll see total execution time for any workflow—plus token usage for AI workflows to help monitor performance. And if you're debugging across multiple screens? Just pop the logs out and drag them wherever you’d like.
⚙️What it does
- Adds a Logs view to the bottom of the canvas that can be opened or collapsed. (Chat also appears here if your workflow uses it).
- Displays a hierarchical list of nodes in the order they were executed—including expanded views of sub-workflows.
- Allows you to click a node in hierarchy to preview inputs and outputs directly, or jump into the full Node Details view with a link.
- Provides ability to toggle input and output data on and off.
- Highlights each node live as it runs, showing when it starts, completes, or fails.
- Includes execution history view to explore past execution data in a similar way.
- Shows roll-up stats like total execution time and total AI tokens used (for AI-enabled workflows).
- Includes a “pop out” button to open the logs as a floating window—perfect for dragging to another screen while debugging.
🛠️How to
To access the expanded logs view, click on the Logs bar at the bottom of the canvas. The view is also opens up when you open the chat window on the bottom of the page.
Insights enhancements for Enterprise#
Two weeks after the launch of Insights, we’re releasing some enhancements designed for enterprise users.
- Expanded time ranges. You can now filter insights over a variety of time periods, from the last 24 hours up to 1 year. Pro users are limited to 7 day and 14 day views.
- Hourly granularity. Drill down into the last 24 hours of production executions with hourly granularity, making it easier to analyze workflows and quickly identify issues.
These updates provide deeper visibility into workflow history, helping you uncover trends over longer periods and detect problems sooner with more precise reporting.
Contributors#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-05-05
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-05-05
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-05-01
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-04-28
This release contains core updates, editor improvements, node updates, and bug fixes.
Breadcrumb view from the canvas#
We’ve added breadcrumb navigation directly on the canvas, so you can quickly navigate to any of a workflow’s parent folders right from the canvas.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-04-25
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-04-22
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-04-22
This release contains core updates, editor updates, node updates, performance improvements, and bug fixes.
Extended HTTP Request tool functionality#
We’ve brought the full power of the HTTP Request node to the HTTP Request tool in AI workflows. That means your AI Agents now have access to all the advanced configuration options—like Pagination, Batching, Timeout, Redirects, Proxy support, and even cURL import.
This update also includes support for the $fromAI function to dynamically generate the right parameters based on the context of your prompt — making API calls smarter, faster, and more flexible than ever.
How to:
- Open your AI Agent node in the canvas.
- Click the ‘+’ icon to add a new tool connection.
- In the Tools panel, select HTTP Request Tool.
- Configure it just like you would a regular HTTP Request node — including advanced options
👉 Learn more about configuring the HTTP Request tool.
Scoped API keys#
Users on the Enterprise plan can now create API keys with specific scopes to control exactly what each key can access.
Previously, API keys had full read/write access across all endpoints. While sometimes necessary, this level of access can be excessive and too powerful for most use cases. Scoped API keys allow you to limit access to only the resources and actions a service or user actually needs.
What’s new
When creating a new API key, you can now:
- Select whether the key has read, write, or both types of access.
- Specify which resources the key can interact with.
Supported scopes include:
- Variables — list, create, delete
- Security audit — generate reports
- Projects — list, create, update, delete
- Executions — list, read, delete
- Credentials — list, create, update, delete, move
- Workflows — list, create, update, delete, move, add/remove tags
Scoped API keys give you more control and security. You can limit access to only what’s needed, making it safer to work with third parties and easier to manage internal API usage.
Drag and Drop in Folders#
Folders just got friendlier. With this release, you can now drag and drop workflows and folders — making it even easier to keep things tidy.
Need to reorganize? Just select a workflow or folder and drag it into another folder or breadcrumb location. It’s a small change that makes a big difference when managing a growing collection of workflows.
📁 Folders are available to all registered users—jump in and get your workspace in order!
Contributors#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-04-16
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-04-15
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-04-14
This release contains API updates, core updates, editor updates, a new node, node updates, and bug fixes.
Insights#
We're rolling out Insights, a new dashboard to monitor how your workflows are performing over time. It's designed to give admins (and owners) better visibility of their most important workflow metrics and help troubleshoot potential issues and improvements.
In this first release, we’re introducing a summary banner, the insights dashboard, and time saved per execution.
1. Summary banner#
A new banner on the overview page that gives instance admins and owners a birds eye view of key metrics over the last 7 days.
Available metrics:
- Total production executions
- Total failed executions
- Failure rate
- Average runtime of all workflows
- Estimated time saved
This overview is designed to help you stay on top of workflow activity at a glance. It is available for all plans and editions.
2. Insights dashboard#
On Pro and Enterprise plans, a new dashboard offers a deeper view into workflow performance and activity.
The dashboard includes:
- Total production executions over time, including a comparison of successful and failed executions
- Per-workflow breakdowns of key metrics
- Comparisons with previous periods to help spot changes in usage or behavior
- Runtime average and failure rate over time
3. Time saved per execution#
Within workflow settings, you can now assign a “time saved per execution” value to any workflow. This makes it possible to track the impact of your workflows and make it easier to share this visually with other teams and stakeholders.
This is just the beginning for Insights: the next phase will introduce more advanced filtering and comparisons, custom date ranges, and additional monitoring capabilities.
Node updates#
- We added a credential check for the Salesforce node
- We added SearXNG as a tool for AI agents
You can now search within subfolders, making it easier to find workflows across all folder levels. Just type in the search bar and go.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-04-10
This release contains new features, new nodes, performance improvements, and bug fixes.
Model Context Protocol (MCP) nodes#
MCP aims to standardise how LLMs like Claude, ChatGPT, or Cursor can interact with tools or integrate data for their agents. Many providers - both established or new - are adopting MCP as a standard way to build agentic systems. It is an easy way to either expose your own app as a server, making capabilities available to a model as tools, or as a client that can call on tools outside of your own system.
While it’s still early in the development process, we want to give you access to our new MCP nodes. This will help us understand your requirements better and will also let us converge on a great general solution quicker.
We are adding two new nodes:
- a MCP Server Trigger for any workflow
- a MCP Client Tool for the AI Agent
The MCP Server Trigger turns n8n into an MCP server, providing n8n tools to models running outside of n8n. You can run multiple MCP servers from your n8n instance. The MCP Client Tool connects LLMs - and other intelligent agents - to any MCP-enabled service through a single interface.
Max from our DevRel team created an official walkthrough for you to get started:
MCP Server Trigger#
The MCP Server Trigger turns n8n into an MCP server, providing n8n tools to models running outside of n8n. The node acts as an entry point into n8n for MCP clients. It operates by exposing a URL that MCP clients can interact with to access n8n tools. This means your n8n workflows and integrations are now available to models run elsewhere. Pretty neat.
Explore the MCP Server Trigger docs
MCP Client Tool#
The MCP Client Tool node is a MCP client, allowing you to use the tools exposed by an external MCP server. You can connect the MCP Client Tool node to your models to call external tools with n8n agents. In this regard it is similar to using a n8n tool with your AI agent. One advantage is that the MCP Client Tool can access multiple tools on the MCP server at once, keeping your canvas cleaner and easier to understand.
Node updates#
- Added a node for Azure Cosmos DB
- Added a node for Milvus Vector Store
- Updated the Email Trigger (IMAP) node
Contributors#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-04-09
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-04-09
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-04-08
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-04-07
This release contains new nodes, node updates, API updates, core updates, editor updates, and bug fixes.
Contributors#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-03-31
This release contains API updates, core updates, editor improvements, node updates, and bug fixes.
Contributors#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-03-27
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-03-27
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-03-26
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-03-26
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-03-25
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-03-25
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-03-24
This release contains a new node, a new credential, core updates, editor updates, node updates, and bug fixes.
Folders#
What can we say about folders? Well, they’re super handy for categorizing just about everything and they’re finally available for your n8n workflows. Tidy up your workspace with unlimited folders and nested folders. Search for workflows within folders. It’s one of the ways we’re making it easier to organize your n8n instances more effectively.
How to use it:
Create and manage folders within your personal space or within projects. You can also create workflows from within a folder. You may need to restart your instance in order to activate folders.
Folders are available for all registered users so get started with decluttering your workspace now and look for more features (like drag and drop) to organize your instances soon.
Enhancements to Form Trigger Node#
Recent updates to the Form Trigger node have made it a more powerful tool for building business solutions. These enhancements provide more flexibility and customization, enabling teams to create visually engaging and highly functional workflows with forms.
- HTML customization: Add custom HTML to forms, including embedded images and videos, for richer user experiences.
- Custom CSS support: Apply custom styles to user-facing components to align forms with your brand’s look and feel. Adjust fonts, colors, and spacing for a seamless visual identity.
- Form previews: Your form’s description and title will pull into previews of your form when sharing on social media or messaging apps, providing a more polished look.
- Hidden fields: Use query parameters to add hidden fields, allowing you to pass data—such as a referral source—without exposing it to the user.
- New responses options: Respond to user submissions in multiple ways including text, HTML, or a downloadable file (binary format). This enables forms to display rich webpages or deliver digital assets such as dynamically generated invoices or personalized certificates.
These improvements elevate the Form Trigger node beyond a simple workflow trigger, transforming it into a powerful tool for addressing use cases from data collection and order processing to custom content creation.
Contributors#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-03-18
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-03-17
This release contains a new node, node updates, editor updates, and bug fixes.
Contributors#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-03-14
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-03-14
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-03-13
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-03-12
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-03-12
This release contains bug fixes and an editor update.
Schema Preview#
Schema Preview lets you view and work with a node’s expected output without executing it or adding credentials, keeping you in flow while building.
- See expected node outputs instantly. View schemas for over 100+ nodes to help you design workflows efficiently without extra steps.
- Define workflow logic first, take care of credentials later. Build your end-to-end workflow without getting sidetracked by credential setup.
- Avoid unwanted executions when building. Prevent unnecessary API calls, unwanted data changes, or potential third-party service costs by viewing outputs without executing nodes.
How to use it:
- Add a node with Schema Preview support to your workflow.
- Open the next node in the sequence - Schema Preview data appears in the Node Editor where you would typically find it in the Schema View.
- Use Schema Preview fields just like other schema data - drag and drop them into parameters and settings as needed.
Don’t forget to add the required credentials before putting your workflow into production.
Contributors#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-03-12
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-03-04
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-03-03
This release contains core updates, editor updates, new nodes, node updates, new credentials, credential updates, and bug fixes.
Tidy up#
Tidy up instantly aligns nodes, centers stickies, untangles connections, and brings structure to your workflows. Whether you're preparing to share a workflow or just want to improve readability, this feature saves you time and makes your logic easier to follow. Clean, well-organized workflows aren't just nicer to look at—they’re also quicker to understand.
How to:
Open the workflow you want to tidy, then choose one of these options:
- Click the Tidy up button in the bottom-left corner of the canvas (it looks like a broom 🧹)
- Press Shift + Alt + T on your keyboard
- Right-click anywhere on the canvas and select Tidy up workflow
Want to tidy up just part of your workflow? Select the specific nodes you want to clean up first - Tidy up will only adjust those, along with any stickies behind them.
Multiple API keys#
n8n now supports multiple API keys, allowing users to generate and manage separate keys for different workflows or integrations. This improves security by enabling easier key rotation and isolation of credentials. Future updates will introduce more granular controls.
Contributors#
Rostammahabadi
Lanhild
matthiez
feelgood-interface
adina-hub
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-03-03
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-03-03
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-28
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-28
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-27
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-27
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-24
This release contains bug fixes, a core update, editor improvements, and a node update.
Improved partial executions#
The new execution engine for partial executions ensures that testing parts of a workflow in the builder closely mirrors production behaviour. This makes iterating with updated run-data faster and more reliable, particularly for complex workflows.
Before, user would test parts of a workflow in the builder that didn't consistently reflect production behaviour, leading to unexpected results during development.
This update aligns workflow execution in the builder with production behavior.
Here is an example for loops:
Before
After
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-21
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-21
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-21
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-21
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-20
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-20
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-17
This release contains bug fixes and an editor improvement.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-17
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-17
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-15
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-15
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-15
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-15
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-12
This release contains new features, node updates, and bug fixes.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-06
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-05
This release contains new features, node updates, and bug fixes.
Contributors#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-04
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-04
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-03
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-02-03
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-01-29
This release contains new features, editor updates, new nodes, new credentials, node updates, and bug fixes.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-01-23
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-01-22
This release contains new features, editor updates, new credentials, node improvements, and bug fixes.
Contributors#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-01-17
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-01-17
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-01-17
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-01-17
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-01-15
This release contains bug fixes and editor updates.
Improved consistency across environments#
We added new UX and automatic changes improvements resulting in a better consistency between your staging and production instances.
Previously, users faced issues like:
- Lack of visibility into required credential updates when pulling changes
- Incomplete synchronization, where changes — such as deletions — weren’t always applied across environments
- Confusing commit process, making it unclear what was being pushed or pulled
We addressed these by:
- Clearly indicating required credential updates when pulling changes
- Ensuring deletions and other modifications sync correctly across environments
- Improving commit selection to provide better visibility into what’s being pushed
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-01-09
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2025-01-08
This release contains new features, a new node, node updates, performance improvements and bug fixes.
Overhauled Code node editing experience#
We added a ton of new helpers to the Code node, making edits of your code much faster and more comfortable. You get:
- TypeScript autocomplete
- TypeScript linting
- TypeScript hover tips
- Search and replace
- New keyboard shortcuts based on the VSCode keymap
- Auto-formatting using prettier (Alt+Shift+F)
- Remember folded regions and history after refresh
- Multi cursor
- Type function in the Code node using JSDoc types
- Drag and drop for all Code node modes
- Indentation markers
We build this on a web worker architecture so you won't have to suffer from performance degradation while typing.
To get the full picture, check out our Studio update with Max and Elias, where they discuss and demo the new editing experience. 👇
New node: Microsoft Entra ID#
Microsoft Entra ID (formerly known as Microsoft Azure Active Directory or Azure AD) is used for cloud-based identity and access management. The new node supports a wide range of Microsoft Entra ID features, which includes creating, getting, updating, and deleting users and groups, as well as adding users to and removing them from groups.
Node updates#
- AI Agent: Vector stores can now be directly used as tools for the agent
- Code: Tons of new speed and convenience features, see above for details
- Google Vertex Chat: Added option to specify the GCP region for the Google API credentials
- HighLevel: Added support for calendar items
We also added a custom projects icon selector on top of the available emojis. Pretty!
Contributors#
igatanasov
Stamsy
feelgood-interface
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-12-19
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-12-19
This release contains node updates, performance improvements, and bug fixes.
Node updates#
- AI Agent: Updated descriptions for Chat Trigger options
- Facebook Graph API: Updated for API v21.0
- Gmail: Added two new options for the
Send and waitoperation, free text and custom form - Linear Trigger: Added support for admin scope
- MailerLite: Now supports the new API
- Slack: Added two new options for the
Send and waitoperation, free text and custom form
We also added credential support for SolarWinds IPAM and SolarWinds Observability.
Last, but not least, we improved the schema view performance in the node details view by 90% and added drag and drop re-ordering to parameters. This comes in very handy in the If or Edit Fields nodes.
Contributors#
CodeShakingSheep
mickaelandrieu
Stamsy
pbdco
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-12-12
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-12-12
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-12-11
This release contains node updates, usability improvements, and bug fixes.
Node updates#
- AI Transform: The
maximum context lengtherror now retries with reduced payload size - Redis: Added support for
continue on fail
Improved commit modal#
We added filters and text search to the commit modal when working with Environments. This will make committing easier as we provide more information and better visibility. Environments are available on the Enterprise plan.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-12-10
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-12-10
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-12-06
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-12-05
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-12-04
This release contains node updates, performance improvements, and bug fixes.
Task runners for the Code node in public beta#
We're introducing a significant performance upgrade to the Code node with our new Task runner system. This enhancement moves JavaScript code execution to a separate process, improving your workflow execution speed while adding better isolation.
Our benchmarks show up to 6x improvement in workflow executions using Code nodes - from approximately 6 to 35 executions per second. All these improvements happen under the hood, keeping your Code node experience exactly the same.
The Task runner comes in two modes:
- Internal mode (default): Perfect for getting started, automatically managing task runners as child processes
- External mode: For advanced hosting scenarios requiring maximum isolation and security
Currently, this feature is opt-in and can be enabled using environment variables. Once stable, it will become the default execution method for Code nodes.
To start using Task runners today, check out the docs.
Node updates#
- AI Transform node: We improved the prompt for code generation to transform data
- Code node: We added a warning if
pairedItemis absent or could not be auto mapped
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-12-04
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-11-29
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-11-27
This release contains node updates, performance improvements and bug fixes.
New canvas in beta#
The new canvas is now the default setting for all users. It should bring significant performance improvements and adds a handy minimap. As it is still a beta version you can still revert to the previous version with the three dot menu.
We're looking forward to your feedback. Should you encounter a bug, you will find a handy button to create an issue at the bottom of the new canvas as well.
Node updates#
- We added credential support for Zabbix to the HTTP request node
- We added new OAuth2 credentials for Microsoft SharePoint
- The Slack node now uses markdown for the approval message when using the
Send and Wait for Approvaloperation
Contributors#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-11-26
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-11-26
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-11-25
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-11-20
This release contains a new feature, node improvements and bug fixes.
Sub-workflow debugging#
We made it much easier to debug sub-workflows by improving their accessibility from the parent workflow.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-11-13
This release contains node updates, performance improvements and many bug fixes.
New AI agent canvas chat#
We revamped the chat experience for AI agents on the canvas. A neatly organized view instead of a modal hiding the nodes. You can now see the canvas, chat and logs at the same time when testing your workflow.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-11-07
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-11-06
This release contains node updates and bug fixes.
Node updates#
- AI Transform: Improved usability
- Anthropic Chat Model Node: Added Haiku 3.5 support
- Convert to File: Added delimiter option for writing to CSV
- Gmail Trigger: Added option to filter for draft messages
- Intercom: Credential can now be used in the HTTP Request node
- Rapid7 InsightVM: Added credential support
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-10-31
This release contains performance improvements, a node update and bug fixes.
Node update#
- Anthropic Chat Model: Added support for claude-3-5-sonnet-20241022
We made updates to how projects and workflow ownership are displayed making them easier to understand and navigate.
We further improved the performance logic of partial executions, leading to a smoother and more enjoyable building experience.
New n8n canvas alpha#
We have enabled the alpha version of our new canvas. The canvas is the ‘drawing board’ of the n8n editor, and we’re working on a full rewrite. Your feedback and testing will help us improve it. Read all about it on our community forum.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-10-28
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-10-25
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-10-25
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-10-24
What changed?
Queue polling via the environment variable QUEUE_RECOVERY_INTERVAL has been removed.
When is action necessary?
If you have set QUEUE_RECOVERY_INTERVAL, you can remove it as it no longer has any effect.
This release contains a new features, new nodes, node enhancements, and bug fixes.
New node: n8n Form#
Use the n8n Form node to create user-facing forms with multiple pages. You can add other nodes with custom logic between to process user input. Start the workflow with a n8n Form Trigger.
Additionally you can:
- Set default selections with query parameters
- Define the form with a JSON array of objects
- Show a completion screen and redirect to another URL
Node updates#
New nodes:
- Google Business Profile and Google Business Profile Trigger: Use these to integrate Google Business Profile reviews and posts with your workflows
Enhanced nodes:
- AI Agent: Removed the requirement to add at least one tool
- GitHub: Added workflows as a resource operation
- Structured Output Parser: Added more user-friendly error messages
For additional security, we improved how we handle multi-factor authentication, hardened config file permissions and introduced JWT for the public API.
For better performance, we improved how partial executions are handled in loops.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2024-10-24
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-10-21
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-10-16
This release contains a new node, node enhancements, performance improvements and bug fixes.
Enhanced node: Remove Duplicates#
The Remove Duplicates node got a major makeover with the addition of two new operations:
- Remove Items Processed in Previous Executions: Compare items in the current input to items from previous executions and remove duplicates
- Clear Deduplication History: Wipe the memory of items from previous executions.
This makes it easier to only process new items from any data source. For example, you can now more easily poll a Google sheet for new entries by id or remove duplicate orders from the same customer by comparing their order date. The great thing is, you can now do this within and across workflow runs.
New node: Gong#
The new node for Gong allows you to get users and calls to process them further in n8n. Very useful for sales related workflows.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2024-10-15
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-10-15
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-10-15
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-10-11
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-10-11
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-10-11
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-10-11
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-10-09
What changed?
- The worker server used to bind to IPv6 by default. It now binds to IPv4 by default.
- The worker server's
/healthzused to report healthy status based on database and Redis checks. It now reports healthy status regardless of database and Redis status, and the database and Redis checks are part of/healthz/readiness.
When is action necessary?
- If you experience a port conflict error when starting a worker server using its default port, set a different port for the worker server with
QUEUE_HEALTH_CHECK_PORT. - If you are relying on database and Redis checks for worker health status, switch to checking
/healthz/readinessinstead of/healthz.
This release contains new features, node enhancements and bug fixes.
Node updates#
- OpenAI: Added the option to choose between the default memory connector to provide memory to the assistant or to specify a thread ID
- Gmail and Slack: Added custom approval operations to have a human in the loop of a workflow
We have also optimized the worker health checks (see breaking change above).
Each credential now has a seperate url you can link to. This makes sharing much easier.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2024-10-08
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-10-07
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-10-02
This release contains new features, node enhancements and bug fixes.
Skipped 1.62.0
We skipped 1.62.0 and went straight to 1.62.1 with an additional fix.
Additional nodes as tools#
We have made additional nodes usable with the Tools AI Agent node.
Additionally, we have added a $fromAI() placeholder function to use with tools, allowing you to dynamically pass information from the models to the connected tools. This function works similarly to placeholders used elsewhere in n8n.
Both of these new features enable you to build even more powerful AI agents by drawing directly from the apps your business uses. This makes integrating LLMs into your business processes even easier than before.
Node updates#
- Google BigQuery: Added option to return numeric values as integers and not strings
- HTTP Request: Added credential support for Sysdig
- Invoice Ninja: Additional query params for getAll requests
- Question and Answer Chain: Added the option to use a custom prompt
Drag and drop insertion on cursor position from schema view is now also enabled for code, SQL and Html fields in nodes.
Customers with an enterprise license can now rate, tag and highlight execution data in the executions view. To use highlighting, add an Execution Data Node (or Code node) to the workflow to set custom executions data.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
Benjamin Roedell
CodeShakingSheep
manuelbcd
Miguel Prytoluk
[email protected]#
View the commits for this version.
Release date: 2024-09-25
This release contains new features, node enhancements and bug fixes.
Node updates#
- Brandfetch: Updated to use the new API
- Slack: Made adding or removing the workflow link to a message easier
Big datasets now render faster thanks to virtual scrolling and execution annotations are harder to delete.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-09-20
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-09-20
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-09-18
This release contains new features, node enhancements and bug fixes.
Queue metrics for workers#
You can now expose and consume metrics from your workers. The worker instances have the same metrics available as the main instance(s) and can be configured with environment variables.
You can now customize the maximum file size when uploading files within forms to webhooks. The environment variable to set for this is N8N_FORMDATA_FILE_SIZE_MAX. The default setting is 200MiB.
Node updates#
Enhanced nodes:
- Invoice Ninja: Added actions for bank transactions
- OpenAI: Added O1 models to the model select
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2024-09-18
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-09-17
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-09-16
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-09-12
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-09-11
Chat Trigger
If you are using the Chat Trigger in "Embedded Chat" mode, with authentication turned on, you could see errors connecting to n8n if the authentication on the sending/embedded side is mis-configured.
This release contains bug fixes and feature enhancements.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2024-09-06
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-09-05
This release contains new features, bug fixes and feature enhancements.
New node: PGVector Vector Store#
This release adds the PGVector Vector Store node. Use this node to interact with the PGVector tables in your PostgreSQL database. You can insert, get, and retrieve documents from a vector table to provide them to a retriever connected to a chain.
See active collaborators on workflows#
We added collaborator avatars back to the workflow canvas. You will see other users who are active on the workflow, preventing you from overriding each other's work.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-08-28
This release contains new features and bug fixes.
Improved execution queue handling#
We are exposing new execution queue metrics to give users more visibility of the queue length. This helps to inform decisions on horizontal scaling, based on queue status. We have also made querying executions faster.
New credentials for the HTTP Request node#
We added credential support for Datadog, Dynatrace, Elastic Security, Filescan, Iris, and Malcore to the HTTP Request node making it easier to use existing credentials.
We also made it easier to select workflows as tools when working with AI agents by implementing a new workflow selector parameter type.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2024-08-26
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-08-23
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-08-21
This release contains node updates, security and bug fixes.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-08-16
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-08-16
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-08-15
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-08-15
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-08-15
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-08-14
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-08-14
The N8N_BLOCK_FILE_ACCESS_TO_N8N_FILES environment variable now also blocks access to n8n's static cache directory at ~/.cache/n8n/public.
If you are writing to or reading from a file at n8n's static cache directory via a node, e.g. Read/Write Files from Disk, please update your node to use a different path.
This release contains a new feature, a new node, a node update and bug fixes.
Override the npm registry#
This release adds the option to override the npm registry for installing community packages. This is a paid feature.
We now also prevent npm downloading community packages from a compromised npm registry by explicitly using --registry in all npm install commands.
New node: AI Transform#
This release adds the AI Transform node. Use the AI Transform node to generate code snippets based on your prompt. The AI is context-aware, understanding the workflow’s nodes and their data types. The node is only available on Cloud plans.
New node: Okta#
This release adds the Okta node. Use the Okta node to automate work in Okta and integrate Okta with other applications. n8n has built-in support for a wide range of Okta features, which includes creating, updating, and deleting users.
Node updates#
Enhanced node:
This release also adds the new schema view for the expression editor modal.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-08-13
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-08-08
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-08-07
This release contains new features, node enhancements, bug fixes and updates to our API.
API update#
Our public REST API now supports additional operations:
- Create, delete, and edit roles for users
- Create, read, update and delete projects
Find the details in the API reference.
Contributors#
CodeShakingSheep
Javier Ferrer González
Mickaël Andrieu
Oz Weiss
Pemontto
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-08-06
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-08-02
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-07-31
This release contains new features, new nodes, node enhancements, bug fixes and updates to our API.
Added Google Cloud Platform Secrets Manager support#
This release adds Google Cloud Platform Secrets Manager to the list of external secret stores. We already support AWS secrets, Azure Key Vault, Infisical and HashiCorp Vault. External secret stores are available under an enterprise license.
New node: Information Extractor#
This release adds the Information Extractor node. The node is specifically tailored for information extraction tasks. It uses Structured Output Parser under the hood, but provides a simpler way to extract information from text in a structured JSON form.
New node: Sentiment Analysis#
This release adds the Sentiment Analysis node. The node leverages LLMs to analyze and categorize the sentiment of input text. Users can easily integrate this node into their workflows to perform sentiment analysis on text data. The node is flexible enough to handle various use cases, from basic positive/negative classification to more nuanced sentiment categories.
Node updates#
Enhanced nodes:
API update#
Our public REST API now supports additional operations:
- Create, read, and delete for variables
- Filtering workflows by project
- Transferring workflows
Find the details in the API reference.
Contributors#
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-07-31
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-07-26
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-07-26
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-07-25
Prometheus metrics enabled via N8N_METRICS_INCLUDE_DEFAULT_METRICS and N8N_METRICS_INCLUDE_API_ENDPOINTS were fixed to include the default n8n_ prefix.
If you are using Prometheus metrics from these categories and are using a non-empty prefix, please update those metrics to match their new prefixed names.
This release contains new features, node enhancements and bug fixes.
Added Azure Key Vault support#
This release adds Azure Key Vault to the list of external secret stores. We already support AWS secrets, Infisical and HashiCorp Vault and are working on Google Secrets Manager. External secret stores are available under an enterprise license.
Node updates#
Enhanced nodes:
Deprecated nodes:
- OpenAI Model: You can use the OpenAI Chat Model instead
- Google Palm Chat Model: You can use Google Vertex or Gemini instead
- Google Palm Model: You can use Google Vertex or Gemini instead
[email protected]#
View the commits for this version.
Release date: 2024-07-23
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-07-23
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-07-18
This release contains new nodes, node enhancements and bug fixes.
New node: Text Classifier#
This release adds the Text Classifier node.
New node: Postgres Chat Memory#
This release adds the Postgres Chat Memory node.
New node: Google Vertex Chat Model#
This release adds the Google Vertex Chat Model node.
完整的发布详情，请参考 GitHub 上的发布版本。
Node updates#
- Enhanced nodes: Asana
[email protected]#
View the commits for this version.
Release date: 2024-07-16
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-07-10
This release contains node enhancements and bug fixes.
Node updates#
- Enhanced nodes: Chat Trigger, Google Cloud Firestore, Qdrant Vector Store, Splunk, Telegram
- Deprecated node: Orbit (product shut down)
Beta Feature Removal#
The Ask AI beta feature for the HTTP Request node has been removed from this version
Contributors#
Stanley Yoshinori Takamatsu
CodeShakingSheep
jeanpaul
adrian-martinez-onestic
Malki Davis
[email protected]#
View the commits for this version.
Release date: 2024-07-03
This release contains a new node, node enhancements, and bug fixes.
Node updates#
- New node added: Vector Store Tool for the AI Agent
- Enhanced nodes: Zep Cloud Memory, Copper, Embeddings Cohere, GitHub, Merge, Zammad
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
Jochem
KhDu
Nico Weichbrodt
Pavlo Paliychuk
[email protected]#
View the commits for this version.
Release date: 2024-07-03
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-07-03
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-07-01
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-07-01
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-06-27
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-06-27
This release contains bug fixes and feature enhancements.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2024-06-26
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-06-20
Breaking change
Calling $(...).last() (or (...).first() or $(...).all()) without arguments now returns the last item (or first or all items) of the output that connects two nodes. Previously, it returned the item/items of the first output of that node. Refer to the breaking changes log for details.
This release contains bug fixes, feature enhancements, a new node, node enhancements and performance improvements.
完整的发布详情，请参考 GitHub 上的发布版本。
New node: HTTP request tool#
This release adds the HTTP request tool. You can use it with an AI agent as a tool to collect information from a website or API. Refer to the HTTP request tool for details.
Contributors#
Daniel
ekadin-mtc
Eric Francis
Josh Sorenson
Mohammad Alsmadi
Nikolai T. Jensen
n8n-ninja
pebosi
Taylor Hoffmann
[email protected]#
View the commits for this version.
Release date: 2024-06-12
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-06-12
This release contains feature enhancements, node enhancements, and bug fixes.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
Jean Khawand
pemontto
Valentin Coppin
[email protected]#
View the commits for this version.
Release date: 2024-06-12
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-06-10
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-06-06
This release contains new features, node enhancements, and bug fixes.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-06-03
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-05-30
This release contains new features, node enhancements, and bug fixes.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-05-28
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-05-22
This release contains new features, node enhancements, and bug fixes.
Backup recommended
Although this release doesn't include a breaking change, it is a significant update including database migrations. n8n recommends backing up your data before updating to this version.
Credential sharing required for manual executions
Instance owners and admins: you will see changes if you try to manually execute a workflow where the credentials aren't shared with you. Manual workflow executions now use the same permissions checks as production executions, meaning you can't do a manual execution of a workflow if you don't have access to the credentials. Previously, owners and admins could do manual executions without credentials being shared with them. To resolve this, the credential creator needs to share the credential with you.
New feature: Projects#
With projects and roles, you can give your team access to collections of workflows and credentials, rather than having to share each workflow and credential individually. Simultaneously, you tighten security by limiting access to people on the relevant team.
Refer to the RBAC documentation for information on creating projects and using roles.
The number of projects and role types vary depending on your plan. Refer to Pricing for details.
New node: Slack Trigger#
This release adds a trigger node for Slack. Refer to the Slack Trigger documentation for details.
Other highlights#
- Improved memory support for OpenAI assistants.
Rolling back to a previous version#
If you update to this version, then decide you need to role back:
Self-hosted n8n:
- Delete any RBAC projects you created.
- Revert the database migrations using
n8n db:revert.
Cloud: contact [email protected].
Contributors#
Ayato Hayashi
Daniil Zobov
Guilherme Barile
Romain MARTINEAU
[email protected]#
View the commits for this version.
Release date: 2024-05-20
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-05-16
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-05-15
This release contains new features, node enhancements, and bug fixes.
Note that this release removes the AI error debugger. We're working on a new and improved version.
New feature: Tools Agent#
This release adds a new option to the Agent node: the Tools Agent.
This agent has an enhanced ability to work with tools, and can ensure a standard output format. This is now the recommended default agent.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2024-05-08
This release contains new features, node enhancements, and bug fixes.
Note that this release temporarily disables the AI error helper.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2024-05-02
Breaking change
Please note that this version contains a breaking change for instances using a Postgres database. The default value for the DB_POSTGRESDB_USER environment variable was switched from root to postgres. Refer to the breaking changes log for details.
This release contains new features, new nodes, node enhancements, and bug fixes.
New feature: Ask AI in the HTTP node#
You can now ask AI to help create API requests in the HTTP Request node:
- In the HTTP Request node, select Ask AI.
- Enter the Service and Request you want to use. For example, to use the NASA API to get their picture of the day, enter
NASAin Service andget picture of the dayin Request. - Check the parameters: the AI tries to fill them out, but you may still need to adjust or correct the configuration.
Self-hosted users need to enable AI features and provide their own API keys
New node: Groq Chat Model#
This release adds the Groq Chat Model node.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
Alberto Pasqualetto
Bram Kn
CodeShakingSheep
Nicolas-nwb
pemontto
pengqiseven
webk
Yoshino-s
[email protected]#
View the commits for this version.
Release date: 2024-04-25
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-04-25
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-04-25
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-04-24
This release contains new nodes, node enhancements, and bug fixes.
New node: WhatsApp Trigger#
This release adds the WhatsApp Trigger node.
Node enhancement: Multiple methods, one Webhook node#
The Webhook Trigger node can now handle calls to multiple HTTP methods. Refer to the Webhook node documentation for information on enabling this.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2024-04-18
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-04-18
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-04-17
This release contains new nodes, bug fixes, and node enhancements.
New node: Google Gemini Chat Model#
This release adds the Google Gemini Chat Model sub-node.
New node: Embeddings Google Gemini#
This release adds the Google Gemini Embeddings sub-node.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
Chengyou Liu
Francesco Mannino
[email protected]#
View the commits for this version.
Release date: 2024-04-17
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-04-15
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-04-12
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-04-11
Breaking change
Please note that this version contains a breaking change for self-hosted n8n. It removes the --file flag for the execute CLI command. If you have scripts relying on the --file flag, update them to first import the workflow and then execute it using the --id flag. Refer to CLI commands for more information on CLI options.
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-04-11
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-04-10
Breaking change
Please note that this version contains a breaking change for self-hosted n8n. It removes the --file flag for the execute CLI command. If you have scripts relying on the --file flag, update them to first import the workflow and then execute it using the --id flag. Refer to CLI commands for more information on CLI options.
This release contains a new node, improvements to error handling and messaging, node enhancements, and bug fixes.
New node: JWT#
This release adds the JWT core node.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2024-04-04
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-04-03
This release contains new nodes, enhancements and bug fixes.
New node: Salesforce Trigger node#
This release adds the Salesforce Trigger node.
New node: Twilio Trigger node#
This release adds the Twilio Trigger node.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-03-28
This release contains enhancements and bug fixes.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-03-26
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-03-25
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-03-20
This release contains new features, new nodes, and bug fixes.
New node: Microsoft OneDrive Trigger node#
This release adds the Microsoft OneDrive Trigger node. You can now trigger workflows on file and folder creation and update events.
New data transformation functions#
This release introduces new data transformation functions:
String
1 2 3 4 5 6 | |
Number
1 2 | |
Object
1 | |
Array
1 | |
Date & DateTime
1 2 | |
Boolean
1 | |
Contributors#
[email protected]#
View the commits for this version.
Release date: 2024-03-15
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-03-15
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-03-13
This release contains new features, node enhancements, and bug fixes.
Support for Claude 3#
This release adds support for Claude 3 to the Anthropic Chat Model node.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
gumida
Ayato Hayashi
Jordan
MC Naveen
[email protected]#
View the commits for this version.
Release date: 2024-03-07
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-03-07
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-03-06
This release contains new features, node enhancements, performance improvements, and bug fixes.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-03-06
Breaking changes
Please note that this version contains a breaking change. HTTP connections to the editor will fail on domains other than localhost. You can read more about it here.
This is a bug fix release and it contains a breaking change.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-02-28
This release contains new features, new nodes, node enhancements and bug fixes.
New nodes: Microsoft Outlook trigger and Ollama embeddings#
This release adds two new nodes.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-02-23
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-02-21
This release contains new features, node enhancements, and bug fixes.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-02-16
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-02-15
This release contains new features, node enhancements, and bug fixes.
完整的发布详情，请参考 GitHub 上的发布版本。
New features#
OpenAI node overhaul#
This release includes a new version of the OpenAI node, adding more operations, including support for working with assistants.
Other highlights:
- Support for AI events in log streaming.
- Added support for workflow tags in the public API.
Contributors#
[email protected]#
View the commits for this version.
Release date: 2024-02-15
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-02-07
This release contains new features, new nodes, node enhancements and bug fixes.
New nodes: Azure OpenAI chat model and embeddings#
This release adds two new nodes to work with Azure OpenAI in your advanced AI workflows:
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2024-02-02
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-01-31
This release contains new features, node enhancements, and bug fixes.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-01-31
Breaking change
This release removes own mode for self-hosted n8n. You must now use EXECUTIONS_MODE and set to either regular or queue. Refer to Queue mode for information on configuring queue mode.
Skip this release
Please upgrade directly to 1.27.1.
This release contains node enhancements and bug fixes.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-01-24
This release contains new features, node enhancements, and bug fixes.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2024-01-22
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2024-01-17
This release contains a new node, feature improvements, and bug fixes.
New node: Chat Memory Manager#
The Chat Memory Manager node replaces the Chat Messages Retriever node. It manages chat message memories within your AI workflows.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-01-16
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-01-10
This is a bug fix release. It includes important fixes for the HTTP Request and monday.com nodes.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-01-10
This release contains new nodes for advanced AI, node enhancements, new features, performance enhancements, and bug fixes.
Chat trigger#
n8n has created a new Chat Trigger node. The new node provides a chat interface that you can make publicly available, with customization and authentication options.
Mistral Cloud Chat and Embeddings#
This release introduces two new nodes to support Mistral AI:
Contributors#
Anush
Eric Koleda
Mason Geloso
vacitbaydarman
[email protected]#
View the commits for this version.
Release date: 2024-01-09
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2024-01-03
This release contains new nodes, node enhancements, new features, and bug fixes.
New nodes and improved experience for working with files#
This release includes a major overhaul of nodes relating to files (binary data).
There are now three key nodes dedicated to handling binary data files:
- Read/Write Files from Disk to read and write files from/to the machine where n8n is running.
- Convert to File to take input data and output it as a file.
- Extract From File to get data from a binary format and convert it to JSON.
n8n has moved support for iCalendar, PDF, and spreadsheet formats into these nodes, and removed the iCalendar, Read PDF, and Spreadsheet File nodes. There are still standalone nodes for HTML and XML.
New node: Qdrant vector store#
This release adds support for Qdrant with the Qdrant vector store node.
Read n8n's Qdrant vector store node documentation
Contributors#
Aaron Gutierrez
Advaith Gundu
Anush
Bin
Nihaal Sangha
[email protected]#
View the commits for this version.
Release date: 2024-01-03
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-12-27
Upgrade directly to 1.22.4
Due to issues with this release, upgrade directly to 1.22.4.
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-12-27
Upgrade directly to 1.22.4
Due to issues with this release, upgrade directly to 1.22.4.
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-12-21
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-12-21
This release contains node enhancements, new features, performance improvements, and bug fixes.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-12-19
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-12-15
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-12-15
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-12-13
This release contains new features and nodes, node enhancements, and bug fixes.
New user role: Admin#
This release introduces a third account type: admin. This role is available on pro and enterprise plans. Admins have similar permissions to instance owners.
New data transformation nodes#
This release replaces the Item Lists node with a collection of nodes for data transformation tasks:
- Aggregate: take separate items, or portions of them, and group them together into individual items.
- Limit: remove items beyond a defined maximum number.
- Remove Duplicates: identify and delete items that are identical across all fields or a subset of fields.
- Sort: organize lists of in a desired ordering, or generate a random selection.
- Split Out: separate a single data item containing a list into multiple items.
- Summarize: aggregate items together, in a manner similar to Excel pivot tables.
Increased sharing permissions for owners and admins#
Instance owners and users with the admin role can now see and share all workflows and credentials. They can't view sensitive credential information.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-12-06
This release contains bug fixes, node enhancements, and ongoing new feature work.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2023-12-05
This is a bug fix release.
Breaking change
This release removes the TensorFlow Embeddings node.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-12-05
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-12-01
Missing ARM v7 support
This version doesn't support ARM v7. n8n is working on fixing this in future releases.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-11-29
Upgrade directly to 1.19.4
Due to issues with this release, upgrade directly to 1.19.4.
This release contains new features, node enhancements, and bug fixes.
LangChain general availability#
This release adds LangChain support to the main n8n version. Refer to LangChain for more information on how to build AI tools in n8n, the new nodes n8n has introduced, and related learning resources.
Show avatars of users working on the same workflow#
This release improves the experience of users collaborating on workflows. You can now see who else is editing at the same time as you.
[email protected]#
View the commits for this version.
Release date: 2023-11-30
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-11-22
This release contains new features and bug fixes.
Template creator hub#
Built a template you want to share? This release introduces the n8n Creator hub. Refer to the creator hub Notion doc for more information on this project.
Node input and output search filter#
Cloud Pro and Enterprise users can now search and filter the input and output data in nodes. Refer to Data filtering for more information.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-11-17
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-11-15
This release contains node enhancements and bug fixes.
Sticky Note Colors#
You can now select background colors for sticky notes.
Discord Node Overhaul#
An overhaul of the Discord node, improving the UI making it easier to configure, improving error handling, and fixing issues.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2023-11-08
This release contains node enhancements and bug fixes.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-11-07
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-11-02
This release contains new features, node enhancements, and bug fixes.
Workflow history#
This release introduces workflow history: view and load previous versions of your workflows.
Workflow history is available in Enterprise n8n, and with limited history for Cloud Pro.
Learn more in the Workflow history documentation.
Dark mode#
Almost in time for Halloween: this release introduces dark mode.
To enable dark mode:
- Select Settings > Personal.
- Under Personalisation, change Theme to Dark theme.
Optional error output for nodes#
All nodes apart from sub-nodes and trigger nodes have a new optional output: Error. Use this to add steps to handle node errors.
Pagination support added to HTTP Request node#
The HTTP Request node now supports an pagination. Read the node docs for information and examples.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2023-10-26
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-10-26
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-10-25
This release contains node enhancements and bug fixes.
Switch node supports more outputs#
The Switch node now supports an unlimited number of outputs.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-10-25
This release contains new features, feature enhancements, and bug fixes.
Upgrade directly to 1.14.0
This release failed to publish to npm. Upgrade directly to 1.14.0.
RSS Feed Trigger node#
This releases introduces a new node, the RSS Feed Trigger. Use this node to start a workflow when a new RSS feed item is published.
Facebook Lead Ads Trigger node#
This releases add another new node, the Facebook Lead Ads Trigger. Use this node to trigger a workflow when you get a new lead.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-10-24
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2023-10-23
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2023-10-23
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2023-10-18
This release contains new features, node enhancements, and bug fixes.
Form Trigger node#
This releases introduces a new node, the n8n Form Trigger. Use this node to start a workflow based on a user submitting a form. It provides a configurable form interface.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2023-10-13
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-10-11
This release contains new features and bug fixes.
External storage for binary files#
Self-hosted users can now use an external service to store binary data. Learn more in External storage.
If you're using n8n Cloud and are interested in this feature, please contact n8n.
Item Lists node supports binary data#
The Item Lists node now supports splitting and concatenating binary data inputs. This means you no longer need to use code to split a collection of files into multiple items.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-10-11
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-10-10
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-10-09
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-10-05
This release contains bug fixes and preparatory work for new features.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-10-04
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
LangChain in n8n (beta)#
Release date: 2023-10-04
This release introduces support for building with LangChain in n8n.
With n8n's LangChain nodes you can build AI-powered functionality within your workflows. The LangChain nodes are configurable, meaning you can choose your preferred agent, LLM, memory, and other components. Alongside the LangChain nodes, you can connect any n8n node as normal: this means you can integrate your LangChain logic with other data sources and services.
Read more:
- This is a beta release, and not yet available in the main product. Follow the instructions in Access LangChain in n8n to try it out. Self-hosted and Cloud options are available.
- Learn how LangChain concepts map to n8n nodes in LangChain concepts in n8n.
- Browse n8n's new Cluster nodes. This is a new set of node types that allows for multiple nodes to work together to configure each other.
[email protected]#
View the commits for this version.
Release date: 2023-09-28
This release contains new features, performance improvements, and bug fixes.
Tournament#
This releases replaces RiotTmpl, the templating language used in expressions, with n8n's own templating language, Tournament. You can now use arrow functions in expressions.
N8N_BINARY_DATA_TTL and EXECUTIONS_DATA_PRUNE_TIMEOUT removed#
The environment variables N8N_BINARY_DATA_TTL and EXECUTIONS_DATA_PRUNE_TIMEOUT no longer have any effect and can be removed. Instead of relying on a TTL system for binary data, n8n cleans up binary data together with executions during pruning.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-09-25
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-09-21
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-09-20
This release contains node enhancements and bug fixes.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-09-14
此版本包含错误修复。
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-09-13
This release contains node enhancements and bug fixes.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2023-09-06
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-09-06
This release contains bug fixes, new features, and node enhancements.
Upgrade directly to 1.6.1
Skip this version and upgrade directly to 1.6.1, which contains essential bug fixes.
TheHive 5#
This release introduces support for TheHive API version 5. This uses a new node and credentials:
N8N_PERSISTED_BINARY_DATA_TTL removed#
The environment variables N8N_PERSISTED_BINARY_DATA_TTL no longer has any effect and can be removed. This legacy flag was originally introduced to support ephemeral executions (see details), which are no longer supported.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-08-31
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-08-31
This release contains new features, node enhancements, and bug fixes.
Upgrade directly to 1.5.1
Skip this version and upgrade directly to 1.5.1, which contains essential bug fixes.
Highlights#
External secrets storage for credentials#
Enterprise-tier accounts can now use external secrets vaults to manage credentials in n8n. This allows you to store credential information securely outside your n8n instance. n8n supports Infisical and HashiCorp Vault.
Refer to External secrets for guidance on enabling and using this feature.
Two-factor authentication#
n8n now supports two-factor authentication (2FA) for self-hosted instances. n8n is working on bringing support to Cloud. Refer to Two-factor authentication for guidance on enabling and using it.
Debug executions#
Users on a paid n8n plan can now load data from previous executions into their current workflow. This is useful when debugging a failed execution.
Refer to Debug executions for guidance on using this feature.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-08-29
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-08-23
This release contains new features, node enhancements, and bug fixes.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2023-08-18
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-08-16
This release contains new features and bug fixes.
Highlights#
Trial feature: AI support in the Code node#
This release introduces limited support for using AI to generate code in the Code node. Initially this feature is only available on Cloud, and will gradually be rolled out, starting with about 20% of users.
Learn how to use the feature, including guidance on writing prompts, in Generate code with ChatGPT.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2023-08-14
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-08-09
This is a bug fix release.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-08-09
This release contains new features, node enhancements, bug fixes, and performance improvements.
Upgrade directly to 1.2.1
When upgrading, skip this release and go directly to 1.2.1.
Highlights#
Credential support for SecOps services#
This release introduces support for setting up credentials in n8n for the following services:
- AlienVault
- Auth0 Management
- Carbon Black API
- Cisco Meraki API
- Cisco Secure Endpoint
- Cisco Umbrella API
- CrowdStrike
- F5 Big-IP
- Fortinet FortiGate
- Hybrid Analysis
- Imperva WAF
- Kibana
- Microsoft Entra ID
- Mist
- Okta
- OpenCTI
- QRadar
- Qualys
- Recorded Future
- Sekoia
- Shuffler
- Trellix ePO
- VirusTotal
- Zscaler ZIA
This makes it easier to do Custom operations with these services, using the HTTP Request node.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-07-27
This is a bug fix release.
Breaking changes
Please note that this version contains breaking changes if upgrading from a 0.x.x version. For full details, refer to the n8n v1.0 migration guide.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-07-26
This release contains new features, bug fixes, and node enhancements.
Breaking changes
Please note that this version contains breaking changes if upgrading from a 0.x.x version. For full details, refer to the n8n v1.0 migration guide.
Highlights#
Source control and environments#
This release introduces source control and environments for enterprise users.
n8n uses Git-based source control to support environments. Linking your n8n instances to a Git repository lets you create multiple n8n environments, backed by Git branches.
Refer to Source control and environments to learn more about the features and set up your environments.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
Adrián Martínez
Alberto Pasqualetto
Marten Steketee
perseus-algol
Sandra Ashipala
ZergRael
[email protected]#
View the commits for this version.
Release date: 2023-07-24
This is a bug fix release.
Breaking changes
Please note that this version contains breaking changes if upgrading from a 0.x.x version. For full details, refer to the n8n v1.0 migration guide.
完整的发布详情，请参考 GitHub 上的发布版本。
[email protected]#
View the commits for this version.
Release date: 2023-07-19
This is a bug fix release.
Breaking changes
Please note that this version contains breaking changes if upgrading from a 0.x.x version. For full details, refer to the n8n v1.0 migration guide.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
Romain Dunand
noctarius aka Christoph Engelbert
[email protected]#
View the commits for this version.
Release date: 2023-07-13
This release contains API enhancements and adds support for sending messages to forum threads in the Telegram node.
Breaking changes
Please note that this version contains breaking changes if upgrading from a 0.x.x version. For full details, refer to the n8n v1.0 migration guide.
完整的发布详情，请参考 GitHub 上的发布版本。
Contributors#
[email protected]#
View the commits for this version.
Release date: 2023-07-05
This is a bug fix release.
Breaking changes
Please note that this version contains breaking changes if upgrading from a 0.x.x version. For full details, refer to the n8n v1.0 migration guide.
Contributors#
[email protected]#
View the commits for this version.
Release date: 2023-07-05
Breaking changes
Please note that this version contains breaking changes. For full details, refer to the n8n v1.0 migration guide.
This is n8n's version one release.
For full details, refer to the n8n v1.0 migration guide.
Highlights#
Python support#
Although JavaScript remains the default language, you can now also select Python as an option in the Code node and even make use of many Python modules. Note that Python is unavailable in Code nodes added to a workflow before v1.0.