Join the conversation

Join the community of Machine Learners and AI enthusiasts.

Sign Up
KseniaseΒ 
posted an update Mar 16, 2025
Post
8419
13+ types of attention mechanisms

Attention mechanisms allow models to dynamically focus on specific parts of their input when performing tasks. In our recent article, we discussed Multi-Head Latent Attention (MLA) in detail and now it's time to summarize other existing types of attention.

Here is a list of 13+ types of attention mechanisms used in AI models:

- Self-attention
- Cross-attention
- Causal attention
- Softmax attention
- Linear attention
- Sliding Window (Local Attention)
- Global attention
- FlashAttention
- Multi-Head Attention (MHA)
- Multi-Query Attention (MQA)
- Grouped-Query Attention (GQA)
- Multi-Head Latent Attention (MLA)
- Interleaved Head Attention (IHA)

Read our post about these core types of attention to learn their specific features and how they work -> 13+ Attention Mechanisms in Transformers and LLMs Explained https://www.turingpost.com/p/attention-types

13+ Attention Mechanisms in Transformers and LLMs Explained -> https://www.turingpost.com/p/attention-types

In this post