After announcing Bedrock last April in preview, Amazon is now making its fully-managed service for generative AI apps generally available.
Along with the GA release, Amazon is also announcing the forthcoming availability of Meta’s Llama 2 13B and 70B parameter models in the next few weeks.
Llama 2 models come with significant improvements over the original Llama models, including being trained on 40% more data and having a longer context length of 4,000 tokens to work with larger documents.
AWS Bedrock includes several foundation models developed at AI21 Labs, Anthropic, Cohere, Stability AI, and Amazon that can be used through an API to create generative AI applications to run on AWS infrastructure.
Amazon Titan Embeddings is a large language model (LLM) that converts text into numerical representations called embeddings to power search, personalization, and Retrieval-Augmented Generation (RAG) use cases.
Retrieval-Augmented Generation is a technique that allows to extend the capability of a model by connecting it to an external knowledge source.
You can explore the capabilities provided by foundation models either using the Amazon Bedrock console, which makes it possible to load example use cases and prompts, or using a playground. Bedrock playgrounds allow you to use a conversational interface to interact with chat, text, and image models and try out different prompt engineering techniques and inference configuration parameters.
Effective prompt engineering is about crafting the perfect query to get the most out of FMs and obtain proper and precise responses. In general, prompts should be simple, straightforward, and avoid ambiguity. You can also provide examples in the prompt or encourage the model to reason through more complex tasks.
AWS Bedrock can use agents to perform specific tasks that connect external data sources to existing APIs, thus enabling the creation of apps that go beyond mere natural language processing. Agents, though, will remain accessible only in preview, along with Amazon Titan Text generation models and Stability AI’s Stable Diffusion XL image generation model.
On a related note, AWS is also extending the capabilities of its CodeWhisperer coding AI by making it possible to leverage a customer’s internal codebase and resources to provide tailored recommendations.
