Table of Content
AI Information
AI Tutorial
AI Agent
Latest Information
Hot Information
LLMs Information
Technical Frontiers
Industry Applications
Prompt Tutorial
AI Programming
AI Video Generation
AI Audio Generation
Knowledge Base Setup
AI Development Guides
Industry Tutorials
Practical Cases
AI Resource Library
Agent Information
Agent Setup
Agent Training

Q Developer has been updated! Your AI development assistant has evolved again!

Written by
Iris Vance
Updated on:June-23rd-2025
Recommendation

The evolution of Amazon Q Developer makes AI development more efficient and smarter.
Core content:
1. Frankfurt region launched to meet EU compliance requirements and reduce latency
2. Full-link Chinese support to improve multi-language development experience
3. /dev Agent version iteration, SWT-Bench and SWE-Bench performance greatly improved

Yang Fangxian
Founder of 53A/Most Valuable Expert of Tencent Cloud (TVP)

Make AI no longer a "toy"

It's the real productivity engine.

Amazon Q Developer continues to evolve

Together with you, we will enter a new era of intelligent development



Upgrading basic service capabilities



01.EU Compliance Support: Frankfurt Region Launched

Amazon Q Developer has added support for the Frankfurt region. For users with R&D and operations teams in the EU, this can meet compliance requirements while providing lower latency responses. When using Amazon Q Developer in the Frankfurt region, data will remain in the EU, and even when inference resources are insufficient, cross-region inference will only use inference resources from other regions in the EU.


02. Full-link Chinese support (Chat/code generation/document parsing)

Whether it is Chat, Inline Chat, /dev Agent or /doc Agent, they all support Chinese input and output.


  • Configuration guide: Add interaction rules (such as "Always speaks in Chinese") in the project's .amazonq /rules / directory

  • Command example: When using /dev Agent, enter "Please reply in Chinese" to trigger Chinese output



03. /dev Agent version iteration

  • Accuracy breakthrough: SWT-Bench test recurrence rate is 49%, and SWE-Bench Verified solution rate is increased to 66% (first on both lists)

  • Applicable scenarios: complex code debugging, multi-module joint debugging and other difficult tasks






04.Enterprise -level Customization Enhancements

  • New support: Dart/Go/Kotlin/PHP/Ruby/Rust/Scala/Bash/PowerShell and other 11 languages

  • Document index: supports Markdown and TEXT files, and design documents can be directly associated with the code base

  • Enterprise value: Generate code that complies with enterprise specifications



05. GitLab Duo with Amazon Q Developer

Official Release

  • Feature highlights: Directly call Q Agent in GitLab Ultimate to complete code development, review and Java version upgrade

  • Scenario coverage: AI-assisted CI/CD process, seamlessly connected to the existing DevOps system



VS Code plugin capability leap



01. Agentic Coding in VSCode (Free for a limited time until June 1st)

After the release of the Agentic coding capability of Amazon Q Developer CLI, it has been well received by developers. This time it has been directly integrated into Visual Studio Code, allowing developers to use Amazon Q's Agent programming capabilities directly in the IDE. Free experience before June 1!


  • Natural language driven file modification and code difference comparison

  • Real-time display of AI reasoning process, supporting multi-round contextual dialogue

  • Two modes: fully automatic execution or manual step-by-step review



02. Contextual control upgrade

  • Precise positioning: specify class/function/global variable as context through @ directive

  • Capacity expansion: Conversation context supports 100K tokens (3 times increase)


03.Other efficiency optimization

  • /test Agent: Added full-language unit test generation capability

  • / Review Agent: Automatically generate security vulnerability repair solutions

  • Persistent conversation: historical context can still be traced after restarting the IDE

  • The plugin supports exporting conversation history into Markdown or HTML format

  • Supports specifying code snippets in PHP, Ruby, Scala, Shell, and Swift as contexts using the @ symbol

  • Code block display optimization: The code block in the dialog panel supports displaying a maximum of 21 lines and supports scrolling within the code block.



Amazon Q Developer CLI

Developer tool innovation



01. MCP protocol support (USB-C in AI, Amazon Q Developer becomes a Lego plug-in in seconds)

  • Cross-tool connection: For example, generate front-end code through Figma MCP, and pull task description through GitHub MCP

  • Enterprise applications: Standardize and connect internal data sources to build customized AI workflows



02. New command system

  • Resource monitoring: /usage to view the usage of the 200K token large context

  • Permission control: /tools trust/untrust fine-grained control of tool permissions

  • Non-interactive mode: --no-interactive can be integrated into the development workflow (e.g. packaged in a container, triggering Q CLI for code review via GitHub Actions)


03. Enhanced productivity

  • Fuzzy search: Ctr l+ K quick positioning command (supports mixed input of Chinese and English)

  • Environment compatibility : configure system proxy via HTTP_PROXY

  • Problem feedback: /issue command to report exceptions with one click



Enterprise-level value summary



  1. Compliance and efficiency: meet EU data sovereignty requirements and reduce the cost of cross-border team collaboration

  2. Local optimization: Chinese support makes demand communication more accurate and reduces misunderstandings

  3. Quality control: unit testing/security scanning automation

  4. Process integration: seamlessly embed into development environments such as GitLab/VS Code, with zero-cost migration





Related PostsView More
How to build your own MCP server
Want to learn how to build your own MCP server? This article will give you a detailed introduction. Through simple steps, you can build powerful and reusable AI functions. The article not only explains the significance of MCP, but also explains why you should use and build an MCP server yourself. It also covers model fine-tuning techniques and methods to enrich your large model knowledge base....
Silas Grey
I too have tried to fine-tune the large model right from the start, until I realized how wrong I was!
Want to make large language models smarter and more practical? This article provides an in-depth analysis for you! It details the two "magic tools" for improving model accuracy: RAG and fine-tuning. RAG is like installing a "real-time search engine" for AI, which is flexible and applicable to a wide range of scenarios; fine-tuning allows AI to "memorize" knowledge directly, which is accurate...
Audrey Miles
It’s over. AI middle platform is shorter-lived than data middle platform
Analyze the rise and fall of AI middle platforms and explore why they are shorter-lived than data middle platforms. In-depth analysis of the key architecture and functions of AI middle platforms, such as capability accumulation and sharing, and full life cycle management. At the same time, reveal its connection with the technical architecture and principles of large models. RAG technology is...
Caleb Hayes