KV cache quantization memory footprint discussion for Qwen3.6 35B A3B

A Reddit user questions whether quantizing KV cache below Q8 is worth the heavy trade-off for Qwen3.6 35B A3B. The post has 31 upvotes and 10 comments discussing memory optimization.