Qwen 3.6 35B-A3B Q4 vs Gemma 4 12B Q8 on 32GB Mac

Reddit user asks which quantized model performs better for coding on a 32GB Mac with 15 tok/s from the Qwen model. Discussion explores trade-offs between model size and quantization level for local use.