Verified75% confidenceFactTime unknown
GQA (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.
3
Sources
75%
Confidence
Long-term
Relevance
6/1/2026
First Seen
Sources
DeepSeek-V3.2-Exp发现RoPE实现Bug：交错格式不匹配导致推理性能下降
twitterdeepseek_ai
Related Entities
Related Claims
VerifiedGrouped 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 reduces the size of the KV Cache by having multiple Query heads share the same set of Key-Value heads.70% similarUnverifiedMQA has all Query heads share the same set of KV, while GQA groups several Query heads together to share one set of KV, serving as a compromise between MHA and MQA.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% similarUnverifiedThe concept of KV Cache groups originates from techniques such as GQA (Grouped Query Attention), where multiple query heads share the same set of KV heads.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/27886API
curl https://kongchang.com/api/v1/knowledge/claims/27886MCP
get_claim(id=27886)