FOUNDATION_MODELS.md

Foundation models

Models used in the workshop:

• Amazon Titan (family)
  • Exclusive to Amazon Bedrock, the Amazon Titan family of models
  • usage: Text generation, Summarization, Semantic search, Image generation, Retrieval Augmented Generation (RAG)
  • models:
    • Amazon Titan Text (LLM)
    • Amazon Titan Embeddings
      • usage: Building Chatbot with Context
    • Amazon Multimodal Embeddings
    • Amazon Titan Image Generator
      • usage: images
• AI21 Jurrasic
  • usage: question answering, text generation, search, and summarization
• Meta Llama 2
  • usage: Meta's fine-tuned LLMs, called Llama-2-Chat, are optimized for dialogue use cases
• Anthropic Claude v2
  • usage: Customer service, Operations, Legal, Insurance, Coding
  • other links:
    • https://www.anthropic.com/news/claude-2
• Stability AI - Stable Diffusion XL - SDXL 1.0
  • usage: generates images of high quality in virtually any art style and is the best open model for photorealism
  • other links:
    • https://stablediffusionxl.com/
    • https://stability.ai/stable-image

Other models not used in workshop (few important)

• OpenAI GPT models
  • OpenAI GPT4-Vision
  • OpenAI DALL-E 2
  • OpenAI Sora
• Midjourney
  • Midjourney - generate images
  • Midjourney Upscalers - upscale images