Rider 2026.1 最新变化

Rider 正不断发展为一个开放式平台，让您能将心仪的 AI 工具引入到专业开发工作流中。

In addition to Junie, Claude Agent, and Codex in the AI chat, you can now choose from a wider range of agents, including GitHub Copilot and Cursor, and dozens of external agents supported via the Agent Client Protocol. 通过新增的 ACP 注册表，您可以发现可用智能体并一键安装。

More improvements for working with AI agents in Rider here.

Rider 2026.1 improves performance in areas you notice during everyday work.

Opening solutions and attaching to running processes is now faster thanks to more efficient indexing of the referenced assemblies. Code completion is also more responsive, particularly for import suggestions, helping you stay in flow with fewer interruptions.

Code an dalysis has also been optimized, contributing to lower overall IDE overhead.

Rider 2026.1 brings improvements to code analysis, new inspections, and support for evolving language features. Enhanced support for C# extension members includes better navigation and refactorings, as well as improved quick-fixes that automatically import missing extension members when needed.

This release also introduces early support for C# 15 Preview language features, such as collection expression arguments and the [ExtendedLayout] attribute, helping you experiment with new syntax as the language evolves. In addition, several new inspections and style options help catch subtle issues, such as short-lived HttpClient instances or incorrect use of ImmutableArray<T> with collection initializers.

Rider’s C# code intelligence is powered by ReSharper. If you want to dive deeper into C# language updates, we describe them in detail on the What’s New in ReSharper 2026.1 page.

Rider's new Smart Step Into for F# support makes it easier for you to choose your step target. The debugger can also display return values for function calls, helping you understand program flow without adding extra logging. Stepping has been improved for async and task expressions, and multiple issues affecting F# debugging have been fixed.

For the full list of changes and fixes, see the F# plugin release notes.

Rider now supports working with single-file C# programs that can be run using dotnet run file.cs. 您可以直接在 IDE 中打开并运行独立的 .cs 文件，无需使用 .csproj 文件。

Rider 的编辑器可以识别包含顶层语句、#! shebangs, and #: directives, including #:package and #:sdk. 代码高亮显示、代码补全、运行标记和调试功能均已提供，让您可以快速创建以单个 C# 文件形式编写的脚本、原型和一次性工具。

Rider 2026.1 brings NuGet’s PowerShell-based workflow into the IDE, offering a Package Manager Console (PMC) that’s currently in preview.

The new console supports standard NuGet PowerShell commands, as well as commands provided by NuGet packages themselves. 完全支持 Entity Framework Core PowerShell 命令，您可以直接在 Rider 中利用熟悉的 EF Core 工作流。

您可以在 NuGet 工具窗口或直接在终端中运行 NuGet 和 Entity Framework Core 命令。 In the NuGet tool window, you can select the package source and default project using UI controls. In the terminal-based console, this context is displayed in the PowerShell prompt and can be changed using PowerShell commands.

Let us know what you think about this feature here.

A new bundled Azure DevOps plugin allows you to clone repositories directly from Rider.

To get started, go to File | Open | Get from Version Control and select Azure DevOps from the list of providers. Rider authenticates you using a personal access token (PAT) and then displays the repositories available to your account so you can clone them directly from the IDE.

You can now view the native disassembly generated from your C# code without leaving Rider. 安装新的 .NET Disassembler 插件后，您可以在新的 ASM Viewer（ASM 查看器）工具窗口中检查 JIT、ReadyToRun 和 NativeAOT 编译器生成的输出。

Developing .NET MAUI applications for iOS from Windows requires connecting Rider to a Mac build host. Rider 2026.1 streamlines this workflow with a new setup experience that automatically prepares the remote environment.

When you connect to a Mac, Rider now verifies that the remote system has everything configured, including macOS, Xcode, the .NET SDK, and required MAUI workloads. If any components are missing, Rider can install or update them automatically, helping you get started faster.

Additionally, you can now build, deploy, and debug MAUI iOS applications from Windows more reliably. We’ve improved diagnostics and connection stability, and fixed issues that previously prevented MAUI apps from deploying to iOS simulators or Windows devices.

Rider 2026.1 fully supports mobile game development for Unreal Engine across both major mobile platforms. 此版本在 Rider 2025.3 引入对 Android 移动设备和 VR 设备调试支持的基础上，新增了对 iOS 移动设备和 VR 设备的支持。

On macOS, you can debug Unreal Engine games running on iOS devices directly from Rider. 设置断点、检查变量、逐步执行代码，并分析调用堆栈 – 一切都可以在熟悉的调试器界面中完成。

根据 Apple 的要求，仍然需要进行初始 Xcode 设置来完成配置和签署。

Inspecting complex variables while debugging Unreal Engine projects is now much faster and more responsive. Expanding containers no longer slows down your debugging session, making it easier to explore object state and understand what’s happening at runtime.

This improvement comes from a rewritten parser and evaluator for Natvis expressions used by Rider’s debugger. Variable inspection with the rewritten evaluator is up to 87 times faster on warm runs and 16 times faster on cold ones, and the debugger’s memory usage has dropped to just over a third of what it was.

Get the full story of how we were able to achieve that in this blog post.

Finding usages, event implementations, and delegate bindings across Unreal Engine Blueprints and C++ code is now more reliable, making it easier to trace how gameplay logic connects across assets.

Code Vision now supports the BlueprintPure specifier and correctly detects blueprint events implementations in Blueprints. Find Usages has also been improved and now identifies additional BlueprintAssignable delegate bindings.

Blueprint usage search now relies on the asset path instead of the Blueprint name, ensuring accurate results even when multiple Blueprints share the same name.

Rider’s C++ intelligence is powered by JetBrains ReSharper C++. For a complete overview of the C++ improvements in this release, see the dedicated What’s New in ReSharper C++ 2026.1 page.

Working with large Unreal Engine codebases is now noticeably faster and more resource-efficient. In our measurements on Epic Games’ Lyra sample project, initial C++ indexing is over 20% faster, helping you get to work sooner when opening a project for the first time.

After opening an already indexed project, the backend uses approximately 14% less memory, reducing the IDE’s footprint and leaving more resources available for the rest of your workflow. Warm startup is also 10% faster, so returning to your project between sessions feels snappier.

Rider now also indexes Unreal Engine plugins by default. With the growing use of plugins in Unreal Engine projects and the recent performance gains, indexing plugins by default provides better code analysis and navigation out of the box. You can restore the previous behavior on the Language & Frameworks | C/C++ | Unreal Engine tab in Settings.

Rider 2026.1 introduces Beta support for CMake-based C++ projects, allowing you to open, edit, and build them directly in the IDE. 这样，您便能更方便地处理跨平台项目，无需切换到其他构建系统。

This initial implementation focuses on C++ workflows, including those for editing and debugging CMake configuration files. 部分工作流和边缘用例仍需优化，我们将在后续版本中持续提升性能和兼容性。 We would love for you to share your experience in this ticket.

This addition is intended to support game development workflows that rely on CMake, such as projects built around CMake-based engines or tooling. For general-purpose C++ and embedded development, CLion remains the primary JetBrains IDE.

Performance investigation for game projects has become a first-class workflow in Rider 2026.1, thanks to the redesigned Unity Profiler integration.

You can now open Unity Profiler snapshots directly in Rider and analyze them in a dedicated Profiler tool window with a structured view of frames and call stacks. A timeline view shows CPU time per frame, helping you quickly identify slow frames and performance hotspots.

Profiler data is tightly integrated with your code. 您可以直接从分析器帧和调用堆栈导航至对应源代码，编辑器装订区域指示器则会内联显示执行时间和内存信息。

This integration reduces context switching between the Unity Editor and Rider, allowing you to move more naturally between performance data and code when investigating runtime issues.

Please note that Rider does not perform profiling itself and relies on profiler data produced by Unity. Learn how to get started from our documentation.

Rider 2026.1 introduces mixed-mode debugging, which lets you debug both managed (.NET and .NET Framework) and native (C/C++) code in a single debugging session. This is particularly useful for applications that cross managed-native boundaries, such as those where .NET code calls into native libraries or game engines built from a mix of managed and native components.

This feature is currently available only on Windows. You can leave your feedback on the feature in this ticket.

Rider’s Godot integration includes a new add-on that bridges the Godot editor and Rider. 在 Godot 工具栏中启用该插件后，它会自动配置 Godot 编辑器设置，让您在 Rider 中获得更加顺畅的开发体验。

在 Godot 编辑器内的 AssetLib 中搜索“JetBrains Rider Integration”即可开始。

Rider 正不断发展为一个开放式平台，让您能将心仪的 AI 工具引入到专业开发工作流中。

随着 AI 智能体的发展，并行运行多项任务已成为大幅节省时间的主要方式，而这正是 Git 工作树最擅长的工作。 To support cutting-edge workflows for AI-boosted software development, Rider now provides first-class support for Git worktrees. 为紧急的修补程序创建一个单独的 工作树，将另一个工作树交给 AI 智能体，并在 main 分支中继续工作 – 这些都可以同时进行，毫不干扰。

即使您不使用智能体，工作树也能节省分支切换的时间，这一优势在大型项目中尤为显著 。

Junie、Codex 和 Claude Agent 的 AI 聊天集成现在对您连接的数据库提供完整的原生支持。 无论是使用 Junie 的自主规划来重构数据层，还是利用 Claude Agent 的推理，您现在都可以直接在 IDE 中使用自然语言查询、分析和修改您的数据库状态。

外部智能体也可以通过 MCP 服务器使用相同的功能。

Data source settings can now be stored in your JetBrains Account via data source templates. 对于 All Products Pack 用户或任何使用多个 JetBrains IDE 实例的用户来说，此项升级尤为实用，您可以在每个支持数据库功能的 JetBrains IDE 中访问数据源模板和设置。

由于我们会继续发展 IDE，并将关注点放在能为开发者创造最大价值的领域，因此我们决定弃用 Code With Me 这一协作式编码和结对编程服务。 近年来，对此类功能的需求已下降，我们将优先采用针对专业软件开发量身定制的现代工作流。

自版本 2026.1 起，Code With Me 将从所有 JetBrains IDE 中解绑。 取而代之的是，它将作为独立插件在 JetBrains Marketplace 上提供。 2026.1 将是正式支持 Code With Me 的最后一个 IDE 版本，因为我们将逐步弃用该服务。

You can read the full announcement and see the sunset timeline in this blog post.