Artificial Analysis has benchmarked the Qwen 3.8 Low and Medium models, reporting strong performance scores.
- The results indicate that the earlier success of these models was not solely due to overthinking.
- The high scores suggest effective optimization in the Qwen 3.8 architecture.