Alibaba (09988) released a preview version of Qwen3.6-Max.

After the release of Qwen3.6-Plus, Alibaba (09988) Qianwen large model team shared an early preview version of the next generation flagship model: Qwen3.6-Max-Preview. Compared to Qwen3.6-Plus, this preview version brings stronger world knowledge and instruction-following capabilities, as well as significantly improved performance in intelligent agent programming on multiple benchmarks. As a preview version, the model is still actively undergoing iterations, with continuous optimization planned for future versions. The main features of Qwen3.6-Max-Preview include significantly improved intelligent agent programming abilities compared to Qwen3.6-Plus, stronger world knowledge and instruction-following, as well as better performance in real-world intelligent agent and knowledge reliability. Model Performance The following shows a comparison of the evaluation of Qwen3.6-Max-Preview with mainstream cutting-edge models. Compared to Qwen3.6-Plus, the preview version achieved significant improvements in intelligent agent programming (such as SkillsBench +9.9, SciCode +10.8, NL2Repo +5.0, Terminal-Bench 2.0 +3.8), stronger world knowledge (SuperGPQA +2.3, QwenChineseBench +5.3), and better instruction-following (ToolcallFormatIFBench +2.8). In the recently updated large model ranking of Artificial Analysis, Qwen3.6-Max-Preview's performance surpassed models such as GLM5.1, MiniMax-M2.7, and others, ranking as the top domestic model. Getting Started with Qwen3.6-Max-Preview Qwen3.6-Max-Preview will be provided through the Alibaba Cloud BaiLian API, with the model name qwen3.6-max-preview. You can also experience it instantly on Qwen Studio. API Updates Alibaba Cloud BaiLian supports industry standard protocols, compatible with OpenAI specification for chat completions and responses API, as well as interfaces compatible with Anthropic. This release supports the preserve_thinking function: preserving all previous rounds of thinking content in messages, recommended for intelligent agent tasks. Summary Qwen3.6-Max-Preview is an early preview version of the next generation flagship model under the Qianwen large model, with significant improvements in intelligent agent programming, world knowledge, and instruction-following compared to Qwen3.6-Plus. It achieved the highest scores in six major programming benchmarks - SWE-bench Pro, Terminal-Bench 2.0, SkillsBench, QwenClawBench, QwenWebBench, and SciCode - with significant improvements compared to its predecessor. It also performed better in knowledge (SuperGPQA, QwenChineseBench) and instruction-following (ToolcallFormatIFBench). As a preview version, Qwen3.6-Max-Preview is still actively in development. The Qianwen large model team will continue to iterate on the model, with future versions bringing further improvements. Currently, the Qianwen 3.6 series has launched Qwen3.6-Max-Preview, Qwen3.6-Plus, Qwen3.6-Flash, and open-sourced Qwen3.6-35B-A3B. Among them, Qwen3.6-Plus topped the daily, weekly, and trending charts of the world's largest three-party API call platform OpenRouter, while Qwen3.6-35B-A3B topped the HuggingFace global open source large model leaderboard. The Qianwen 3.6 series has become the most popular large model among enterprises and developers.