Grouped-Query Attention: Shrinking the KV Cache

Every large language model has to keep a running cache of key and value vectors for every token it has already generated, and in plain multi-head attention that cache grows with the number of attention heads. Grouped-Query Attention (GQA) — used in LLaMA-2/3, Mistral, Gemma and most modern serving stacks — lets many query heads share one key/value head instead of owning their own, cutting the cache size (and the memory bandwidth needed to read it every token) by a large, tunable factor. This simulator renders the head-sharing topology in 3D: an outer ring of query heads, an inner ring of shared KV heads, and live attention-flow particles connecting each group, while you tune the head count, the group count, the cached sequence length and the model depth and watch the real KV-cache-size formula update in real time.