Introducing Qwen 3.6 Max

ChatBotKit now supports Qwen 3.6 Max, giving teams access to Alibaba Cloud's newest premium Qwen model for advanced coding, long-context reasoning, and multimodal agent execution. It is available across bots, blueprints, skillsets, and the API, making it immediately usable anywhere you already build with ChatBotKit.

Qwen 3.6 Max is positioned as a substantial step forward from earlier Qwen Max and Qwen 3.6 Plus releases, with a specific focus on better coding performance, more efficient agent execution, stronger front-end development output, and improved long-tail knowledge retention. In practice, that translates into a model that is better suited to real production work rather than isolated benchmark tasks - especially when an agent needs to stay effective across multiple steps, tools, and inputs.

The model supports a 240K token context window with up to 64K output tokens, and combines reasoning, tool use, image understanding, and file input in one configuration. That mix is particularly valuable for workflows where an assistant needs to interpret screenshots, inspect uploaded files, call actions, and then produce implementation-grade output without switching models mid-flow. For teams building development copilots, internal assistants, and file-aware automations, Qwen 3.6 Max expands what can be handled in a single model session.

Its pricing also places it clearly in the premium tier at $1.30 per million input tokens and $7.80 per million output tokens, which makes the value proposition straightforward: use it when the quality of reasoning, execution reliability, and multimodal workflow coverage matter more than absolute cost minimisation. That is often the right tradeoff for higher-stakes engineering and operational tasks where a stronger first pass saves real time.

Qwen 3.6 Max is available now in the ChatBotKit model picker and through the API.