Unverified50% confidenceFactExact time
GQA reduces the size of the KV Cache by having multiple Query heads share the same set of Key-Value heads.
1
Sources
50%
Confidence
Long-term
Relevance
9/4/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedMQA令所有查询头共享同一组Key-Value以压缩KV Cache78% similarVerifiedGrouped Query Attention (GQA) and Multi-Query Attention (MQA) reduce KV cache memory usage during inference by having multiple query heads share the same group of key/value heads, and are standard optimizations in models such as Llama 3 and Mistral.77% similarUnverifiedGQA (Grouped-Query Attention) and MQA (Multi-Query Attention) were the industry's solutions prior to MLA for alleviating the KV cache problem by sharing Key/Value heads.70% similarUnverifiedGQA让多个查询头共享同一组键值头，Llama 2 70B 和 Llama 3 系列均采用了 GQA70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/854148API
curl https://kongchang.com/api/v1/knowledge/claims/854148MCP
get_claim(id=854148)