Qwen3.5 Max flagship model is here, climbing back to the top as the #1 in China! On the large model competition platform LM Arena, it scored a total of 1464 points, ranking 15th overall. It’s behind Anthropic, Google, xAI, and OpenAI, but has finally reclaimed the top spot among domestic models. Previously, the top domestic model was ByteDance's Seed2.0. By the way, I recently tried doubao-seed-2.0-pro, which is a Kimi2.5, on the same level as GLM5.0. Not much difference, but ByteDance barely did any promotion for it. I’ve hardly seen any AI bloggers reviewing Seed2.0, which just shows that other AI companies have spent a lot of money on these bloggers. In contrast, ByteDance seems to have spent almost nothing on advertising, so you basically don’t see anyone writing comparison reviews.