Just tried out Qwen Code, and it's pretty handy. You get 1,000 uses per day, with a max of 60 per minute, and there's no token limit—basically enough for regular use. All the command-line features come with Chinese annotations. Qwen Code is a secondary development based on the open-source Gemini Code. It currently uses the latest model, Qwen3.6 plus. Its coding model ranks ninth globally, slightly weaker than GLM5.1 (ranked third worldwide). The current Qwen3.6 plus is just a mid-tier model, but the flagship Qwen3.6 max model is expected to surpass GLM5.1. Normally, it should be released about a month after the plus version, so we might see it in May. At the end of April, there's also a new model from DeepSeek. It’ll still be a bit challenging for Qwen's flagship model to claim the top spot domestically. GitHub open-source repository: QwenLM/qwen-code