How DeepSeek V2 Solves the KV Cache Memory Problem with MLA? The DeepSeek team introduced a new approach called Multi-Head Latent Attention (MLA) in their paper for DeepSeek V2, tackling a key bottleneck in LLMs: the size of the Key Value (KV) cache In standard transformer architectures, the KV cache stores the key and value vectors for each token in the input sequence When new tokens are generated, the cache allows the model to efficiently access past information without recomputing it for ever