Since Qwen 3.6 27b outperforms Gemma 4 26b in most benchmarks I'm not sure the value - also Gemma 26b is a MOE model whereas this is a dense model, so not typic

Since Qwen 3.6 27b outperforms Gemma 4 26b in most benchmarks I'm not sure the value - also Gemma 26b is a MOE model whereas this is a dense model, so not typically direct competitors at their sizes - Gemma 4 31b comparison would be interesting though.