:What Is Qoder CN

Code completion

Trained on a massive corpus of high-quality open-source code, Qoder CN generates line-level and function-level code, unit tests, and optimization suggestions based on the current file and cross-file context. With second-level generation speed, it lets you stay in the flow and focus on technical design.

AI chat (Ask mode)

Ask mode draws on a large body of R&D documents, product documentation, and general engineering knowledge, combined with project-level context awareness, to help you solve problems while coding. It can fix code issues, debug, and troubleshoot runtime errors.

File edit (Edit mode)

Edit mode supports multi-file code changes. Based on your requirements and the current project, it modifies multiple files and supports iteration and review, so you can complete code changes efficiently and stay in control.

Agent mode

Agent mode features autonomous decision-making, environment awareness, and tool use. It uses tools such as project search, file editing, and the terminal to deliver coding tasks end to end. You can also configure MCP tools to align coding with your existing workflows.

Model selector

Switch freely between leading Chinese LLMs such as GLM, DeepSeek, Kimi, and MiniMax. Choose the model that best matches your scenario, language, and task complexity to balance capability and cost.

Expert team

Domain-specific expert agents are available for areas such as front-end, back-end, databases, operations, and testing. You can collaborate with these experts on demand to get focused code suggestions, problem diagnosis, and best practices.

Quest 2.0 with Repowiki and Subagent

Task planning and execution are upgraded across the board. Quest 2.0 automatically breaks complex R&D tasks into executable steps and combines Repowiki (project-level knowledge base) and Subagent (sub-agents) to deliver project-level work end to end. Repowiki, Quest, and Subagent are available without usage limits.

Inline code completion

• Real-time line-level and function-level continuation: Generates line-level and function-level code in real time based on the current syntax and cross-file context, with automatic project awareness.

• Code from comments: Describe what you want in a comment and the code is generated directly in the editor, keeping your coding flow uninterrupted.

AI chat

• R&D Q&A: When you run into a coding question or technical issue, invoke Qoder CN with one shortcut to get answers and solutions without leaving the IDE.

• Project Q&A: Use chat to quickly understand the current repository, search code, and describe requirements in natural language to get end-to-end fix suggestions and proposed code for simple requirements or defects.

• Multi-modal Q&A with images: Add images to the context by selecting, dragging, or pasting them. Qoder CN analyzes the image content and generates code suggestions or fix proposals based on your requirement.

• Enterprise knowledge base Q&A: Use enterprise knowledge and data for chat to quickly build an internal R&D knowledge assistant and improve team efficiency and collaboration.

File edit

• Project-level changes: Modify multiple files in the project based on your task description, iterate through multiple conversations, and roll back via snapshots, so you and Qoder CN can complete coding tasks together step by step.

• Precise editing: Code changes stay within the context you provide, with no surprising modifications outside that scope.

• Fast execution: Strictly follows your task description and provided context to modify files, without elaborate task planning. This makes file edit faster than agent mode for the same change.

• Tool use: Built-in tools include file reading, project-level semantic search, and file editing, so you can complete code changes quickly.

Coding agent

• Project-level changes: Based on your task description, the agent autonomously breaks down the task and modifies multiple files. You can iterate through multiple conversations or roll back via snapshots, and complete coding tasks together with Qoder CN.

• Automatic project context awareness: The agent picks up project information such as the framework, technology stack, required files, and error messages from your task description, without manual context setup.

• Tool use: The agent autonomously uses more than ten built-in coding tools, such as file read/write, code search, and error troubleshooting. It also automatically detects and uses MCP tools.

• Terminal command execution: While running a coding task, the agent decides which commands are needed, writes them, and runs them in the terminal automatically, which significantly speeds up task execution.

Enterprise management

• Centralized access management: Grant and manage access for developers across your organization from a single console.

• Usage analytics: View statistics such as developer activity, the share of AI-generated code, and chat usage to quickly assess adoption and impact.

• Knowledge management: Manage enterprise private data efficiently and combine it with Qoder CN to provide enterprise knowledge base Q&A and code generation enhancement that fit your business scenarios.

• Custom extensions: Combine multiple context sources such as code changes, files, and inputs to define custom commands for common enterprise scenarios, and extend coding workflows flexibly.

• Dedicated VPC deployment: The Enterprise Dedicated Edition supports dedicated VPC deployment and connections to your internal network, which protects enterprise data end to end and meets security and compliance requirements.