Preview. These maps are our reading of public sources and haven't been checked with the projects' maintainers yet.
Qwen Code
Alibaba Qwen team's terminal coding agent, forked from Gemini CLI v0.8.2 and developed independently since; multi-protocol (OpenAI/Anthropic/Gemini/Qwen APIs), local models, IDE plugins via ACP, desktop and web UI.
QwenLM/qwen-code v0.26.1 released 2026-10-24; repo pushed 2026-10-25. Free Qwen OAuth tier discontinued 2026-04-15. (checked 2026-10-25)
In Coding agents
Where it stands apart
- How is it mainly sustained? big company side project (10% of 21 others)
- Out of the box, does it ask you before editing files or running shell commands? edits without asking, asks before most commands (14% of 21 others)
What it shares with most of the community
- Is it actively maintained? active (71% of 21 others)
- How is the main thing licensed? permissive open source (MIT/Apache/BSD) (52% of 21 others)
- Can it run entirely on your own device or infrastructure, without the maker's servers? yes (52% of 21 others)
- Who is it for? developers (95% of 21 others)
- How does finished work usually come back to you? edits in your local working tree while you watch (52% of 21 others)
- Whose models can you use? Pick the first that applies. any provider or endpoint you configure (your own key, custom URL, or local model) (67% of 21 others)
- Is pointing it at a model running on your own machine (e.g. Ollama, LM Studio) a documented setup? yes, documented (67% of 21 others)
- Where do the commands it runs execute by default? on your machine with your user's permissions (86% of 21 others)
- Can users add their own tools through MCP (Model Context Protocol) servers? yes (95% of 21 others)
- Does it read project instructions from AGENTS.md? yes, alongside its own file (52% of 21 others)
- When you first set it up, how do you pay for the models it uses by default? with your own API key or local model (the tool itself is free) (52% of 21 others)
Nearest neighbors
- Cline 82% of 17 shared answers match
- Crush 76% of 17 shared answers match
- OpenCode 76% of 17 shared answers match
Every answer
| Question | Answer | Source |
|---|---|---|
| Is it actively maintained? | active | source 2026-10-25 |
| How is the main thing licensed? | permissive open source (MIT/Apache/BSD) Apache-2.0. | source 2026-10-25 |
| How is it mainly sustained? | stands apart big company side project Alibaba's Qwen team; promotes Alibaba Cloud Coding/Token Plans. | source 2026-10-25 |
| Can it run entirely on your own device or infrastructure, without the maker's servers? | yes Open-source agent plus open-weight Qwen models via Ollama/vLLM. | source 2026-10-25 |
| Do you need an account to use it? | optional account Third-party and custom providers work; QwenCloud key optional; free OAuth ended. | source 2026-10-26 |
| Who is it for? | developers | source 2026-10-25 |
| Which shared standard or format does it mainly build on? | MCP (Model Context Protocol) MCP, ACP and AGENTS.md; QWEN.md is its own file. Low confidence. | source 2026-10-26 |
| Where do you primarily use it? | terminal (CLI or TUI) Also IDE plugins, desktop app, web UI, chat integrations. | source 2026-10-25 |
| How does finished work usually come back to you? | edits in your local working tree while you watch | source 2026-10-25 |
| Whose models can you use? Pick the first that applies. | any provider or endpoint you configure (your own key, custom URL, or local model) 'Any third-party provider or local model'; Custom Provider for OpenAI/Anthropic/Gemini-compatible endpoints. | source 2026-10-25 |
| Is pointing it at a model running on your own machine (e.g. Ollama, LM Studio) a documented setup? | yes, documented Ollama / vLLM. | source 2026-10-25 |
| Out of the box, does it ask you before editing files or running shell commands? | stands apart edits without asking, asks before most commands POLL GAP: default tools.approvalMode is 'auto', where an LLM classifier auto-approves safe actions and blocks risky ones. No option fits a classifier mode; closest chosen. | source 2026-10-25 |
| Where do the commands it runs execute by default? | on your machine with your user's permissions tools.sandbox default undefined; Docker/Podman or Linux bwrap/landlock sandbox opt-in. | source 2026-10-25 |
| Can users add their own tools through MCP (Model Context Protocol) servers? | yes | source 2026-10-25 |
| Does it read project instructions from AGENTS.md? | yes, alongside its own file Defaults to both QWEN.md and AGENTS.md, QWEN.md first. | source 2026-10-25 |
| Does it use the Agent Client Protocol (ACP) for agent-editor connections? | yes, as an agent that editors can drive Zed/JetBrains integrations; 'qwen serve' daemon speaks ACP. | source 2026-10-25 |
| When you first set it up, how do you pay for the models it uses by default? | with your own API key or local model (the tool itself is free) Free OAuth tier ended 2026-04-15; users bring API keys or buy Alibaba Cloud Coding Plan. | source 2026-10-25 |
Sources
This page was derived from:
- GitHub · github.com/QwenLM/qwen-code · read 2026-10-25
- GitHub · github.com/QwenLM/qwen-code/blob/main/packages/core/src/utils/memory-constants.ts · read 2026-10-25
- GitHub · github.com/QwenLM/qwen-code/releases · read 2026-10-25
- GitHub · raw.githubusercontent.com/QwenLM/qwen-code/main/README.md · read 2026-10-25
- Web page · qwenlm.github.io/qwen-code-docs/en/users/configuration/auth/ · read 2026-10-25
- Web page · qwenlm.github.io/qwen-code-docs/en/users/configuration/settings/ · read 2026-10-25
RSS for this project · Its own release feed
Are you a maintainer? Claiming and correcting projects is coming soon.