Model17h ago
Alibaba Qwen3.6 Outperforms Google Gemma 4
The Qwen3.6-35B-A3B model activates only three billion parameters per token. Despite this efficiency, it beats Google's Gemma 4-31B on agentic coding and reasoning benchmarks. This proves that sparse architectures can outperform denser models with fewer active weights. Developers now have a more efficient open-source option for complex coding tasks.