[magicalhippo]

The Qwen3.5 27B model did almost the same as Sonnet 4.5 in this[1] reasoning benchmark, results here[2].

Obviously there's more to a model than that but it's a data point.

[1]: https://github.com/fairydreaming/lineage-bench

[2]: https://github.com/fairydreaming/lineage-bench-results/tree/...