Alibaba opens the next-generation flagship Qwen3.6-Max-Preview for preview, with a focus on intelligent agent programming

ME News: April 20 (UTC+8). According to Beating monitoring, the Alibaba Qianwen team released Qwen3.6-Max-Preview, positioned as an early preview of the next-generation flagship model, replacing the existing Qwen3.6-Plus. Users can chat and experience it directly in Qwen Studio (chat.qwen.ai). They can then access the API via Alibaba Cloud Bailian by opening it with the model name qwen3.6-max-preview. The interface is compatible with OpenAI’s chat completions and responses standards, as well as Anthropic’s protocols.

This version is mainly aimed at agentic coding, enabling the model to write code, run it, view errors, call tools, and complete multi-step programming tasks like a programmer. Compared with the previous Qwen3.6-Plus, the official said the improvements are concentrated in programming: SkillsBench +9.9, SciCode +10.8, NL2Repo +5.0, and Terminal-Bench 2.0 +3.8 points. Gains in world knowledge and tool-calling formats are reported to follow, with improvements ranging from 2.3 to 5.3 points across three other metrics.

The official claims that it achieved the highest scores across six programming benchmarks, including SWE-bench Pro, Terminal-Bench 2.0, and SciCode. From the naming, QwenClawBench and QwenWebBench appear to be Qianwen’s own evaluation sets, which should be considered separately from public benchmarks.

On the API side, a new preserve_thinking option has also been added. When enabled, messages will retain the thinking content from the previous few rounds. In reasoning models, the default behavior is to return only the “thinking of the current round” once per turn. In multi-round agent conversations, the context lacks the thinking from earlier steps, so when the model replans it may go back to previously tried paths or forget what it has already tried. This toggle addresses that gap.

(Source: BlockBeats)