Evaluating a Proposed Attention Change for Cost Reasons

Your infra team proposes fine-tuning your self-hosted model to use a more aggressive grouped-query attention configuration — going from 16 KV heads down to 4 — specifically to cut KV cache memory and let you raise the concurrent-request limit on your current GPUs.

1. Explain what's actually changing in the attention mechanism, and why it reduces KV cache size the way it does.
2. Does this change affect the query-side computation or only the key/value side? Why does that distinction matter for the trade-off you're evaluating?
3. What would you want to check before signing off on this change, beyond "the cache got smaller"?