# Overview A type of [[Deep Learning]] model used for [[Natural Language Processing (NLP)]] which is trained on a very large set of text data with the goal of generating language. For comparison, reference [Countless.dev \| AI Model Comparison](https://countless.dev/). # Key Considerations ## Types of LLMs - Base LLM - predicts next word, based on text training data - Instruction Tuned LLM - fine-tune on instructions and good attempts at following those instructions ## List of LLMs - Closed Source - [[OpenAI]] - [[GPT-4 and 4o]] - [[GPT-3]] - [[Anthropic]] - [[Claude Sonnet]] - [[Claude Haiku]] - [[Claude Opus]] - [[Google]] - [[Gemini]] - Open Source - [[LLaMA 3]] - [[BERT]] - [[T5]] ## Common Use Cases - Text Summarization - Language Translation - Question Answering - Search Augmentation and Information Retrieval ([[Retrieval Augmented Generation (RAG)]]) - Text Classification - [[Zero-shot Classification]] - [[Few-shot Classification]] # Implementation Details # Useful Links # Related Topics - Pricing comparisons: [Countless.dev | AI Model Comparison](https://countless.dev/) ## Reference - [Automation Platform v2: Improving Conversational AI at Airbnb \| by Chutian Wang \| The Airbnb Tech Blog \| Medium](https://medium.com/airbnb-engineering/automation-platform-v2-improving-conversational-ai-at-airbnb-d86c9386e0cb) #### Working Notes #### Sources - [Fast LLM Inference From Scratch](https://andrewkchan.dev/posts/yalm.html) #### Related Topics - [[GenAI]]