Released: Jul 16, 2026

Aggiornamenti in 2026.2

Funzionalità

Go performance optimization

• 'pprof' profiling without additional setup - Previously, profiling in GoLand required using a test as the entry point. GoLand 2026.2 extends pprof profiling beyond tests and now supports profiling regular Go applications. You can collect CPU, memory, mutex, block, and goroutine profiles using standard run configurations. You can also import and export pprof profiles. This lets you capture performance data from production environments and analyze it in GoLand.
• Support for the Goroutine leak profile - GoLand 2026.2 adds support for the new Goroutine leak profile introduced in Go 1.27. It helps you find goroutines that are permanently blocked and can never resume execution because the synchronization primitive they are waiting on, such as a channel, sync.Mutex, or sync.Cond, has become unreachable.
• Redesigned profile viewer - GoLand 2026.2 introduces a redesigned profile viewer that makes performance data easier to explore and understand. Switch between multiple visualizations, including flame graphs, call trees, top functions, and graph views. The graph visualization helps you quickly understand relationships between callers and callees, identify hot paths, and navigate directly to the relevant source code.
• Profiling data in the editor gutter - You no longer need to switch between profile views and source code to understand where your application spends execution time. GoLand now visualizes profiling results directly in the editor gutter, showing how much time was spent at specific lines and call sites. You can quickly identify hotspots, navigate to expensive operations, and understand performance bottlenecks in the context of your code.
• Live CPU and memory charts - The new live performance charts show CPU and memory consumption directly in the IDE while your application runs. You can observe resource usage as it changes, verify the impact of optimizations, and quickly spot suspicious spikes without switching to external monitoring tools.
• Struct optimization tool - The order of fields in a struct can affect how much memory it uses. Even small inefficiencies can add up when a struct is created thousands or millions of times. The new memory alignment tool detects opportunities to reduce memory usage by rearranging struct fields without changing your code's behavior. You can apply the suggested optimization with a single quick-fix. This is especially useful for applications that process large amounts of data or keep many objects in memory. You can run the analysis manually or enable the inspection in the editor.
• Escape analysis support - When you need to investigate unexpected memory usage, performance issues, high allocation activity, or increased garbage collection pressure, compiler allocation decisions can provide valuable clues. GoLand now visualizes Go compiler escape analysis directly in the editor, helping you understand why certain values are allocated on the stack or the heap. Instead of parsing compiler output, you can inspect allocation decisions alongside your code.

Modernize your code with confidence

• Support of go fix - GoLand 2026.2 now integrates with the official go fix tool and surfaces its recommendations directly in the editor. When the Go team introduces a new language or standard library migration, GoLand highlights affected code and offers the same quick-fixes provided by the Go toolchain. Apply updates directly in the editor to ensure your codebase stays up to date as new fixes become available. You can review and apply updates without leaving the IDE, making it easier to adopt new Go versions across large projects.

AI

• Agent skills manager - The new skills manager gives you a dedicated place to view and manage agent skills. Skills provide reusable workflow knowledge that helps AI agents understand how to perform specific tasks and interact with available tools. This makes agent behavior more predictable and easier to customize.
• Configurable AI completion models - AI completion is an in-editor code suggestion experience that combines inline completion and next edit suggestions that go beyond the cursor. Available to all JetBrains AI users out of the box, it's powered by JetBrains-trained models and doesn't use up your AI credits. With the latest update, you can now connect your own OpenAI-compatible model providers for AI completion.
• GitHub Copilot built into GoLand - You can now access GitHub Copilot directly from the JetBrains AI chat. Sign in with your GitHub account, select a Copilot model, and use Copilot-powered chat capabilities alongside other supported agents from a unified interface. This removes the need to switch between separate AI integrations inside the IDE.

Application and infrastructure workflows

• Inline service actions in Docker Compose - Working with multicontainer applications often requires switching between configuration files, logs, browsers, and terminal windows. GoLand now provides inline actions directly inside Docker Compose files. You can view service status, access logs, open exposed endpoints, and perform common actions directly from docker-compose.yaml.
• Service templates for Docker Compose - When building a Docker Compose configuration from scratch, you usually have to search documentation and copy service definitions. GoLand now provides smart service templates for common technologies such as PostgreSQL, Kafka, Redis, and others. Start typing a service name and the IDE generates a ready-to-use configuration that you can customize for your environment.
• Terraform testing framework support - GoLand 2026.2 adds support for the Terraform testing framework, bringing code insight, navigation, completion, and validation to Terraform test files. You can now work with infrastructure tests using the same IDE assistance available throughout the rest of your Terraform workflow.

IDE performance and workflows

• Automatic creation of run/debug configurations - GoLand 2026.2 can now automatically detect main packages in your project and create run configurations for them. Instead of manually configuring each application entry point, you can start running, debugging, or profiling your code right away from the gutter or the Run/Debug Configurations menu.
• Faster code analysis for large projects - GoLand 2026.2 improves the performance of inspections, highlighting, and project-wide analysis, reducing the time spent processing large files and complex codebases. These optimizations help keep the IDE responsive while you edit, navigate, and analyze code.
• Improved performance when saving files - GoLand now performs file writes asynchronously, reducing UI freezes when working with large projects, WSL environments, Docker containers, and remote file systems. The result is a smoother editing experience, especially in environments where file operations are slower.
• Git worktree improvements - In GoLand 2026.2, you can use worktrees in WSL environments, remove pruned worktrees directly from the IDE, and benefit from improved cleanup and deletion handling. These improvements make parallel development workflows easier to manage.
• Deprecation of the Machine Learning Code Completion plugin - The Machine Learning Code Completion plugin is now deprecated and no longer bundled with GoLand. As AI completion, next edit suggestions, and AI agents have become the primary ways developers receive coding assistance, usage of ML-based completion ranking has declined. If you still rely on the plugin, you can continue to install and use it from JetBrains Marketplace.