Amazon Web Services (AWS) has made significant strides in the field of generative artificial intelligence (AI) with the release of five groundbreaking innovations. These innovations aim to provide organisations of all sizes with the tools to develop generative AI applications, enhance employee productivity, and drive transformative changes within their operations.
1. Amazon Bedrock: A One-Stop Solution for Foundation Models
One of the standout innovations is the general availability of Amazon Bedrock, a fully managed service designed to simplify access to foundation models (FMs) from leading AI companies. Amazon Bedrock consolidates these FMs into a single application programming interface (API), providing customers with a wide range of choices. This empowers businesses to experiment with different models to find the one that best suits their unique use case.
2. Amazon Titan Embeddings: Expanding Model Selection
AWS has also announced the general availability of Amazon Titan Embeddings, a family of models designed to support various use cases. These models are pre-trained on extensive datasets and are suitable for tasks such as search, personalisation, and retrieval-augmented generation (RAG). These models can be augmented with additional data to enhance their responses, making them invaluable for businesses seeking to leverage generative AI.
3. Llama 2: Meta’s Next-Generation LLM
Llama 2, Meta’s advanced large language model (LLM), will soon be available through Amazon Bedrock, marking a significant milestone. These models have undergone substantial improvements, with training on larger datasets and extended context lengths. Llama 2 models are ideal for dialogue-based use cases, allowing organisations to build generative AI applications without the complexities of managing infrastructure.
4. Amazon CodeWhisperer: Secure Customisation for Developers
AWS is introducing a new capability for Amazon CodeWhisperer, an AI-powered coding companion. This capability enables secure customisation of CodeWhisperer’s code suggestions using an organisation’s private codebase. Developers can save time and enhance productivity by obtaining code recommendations tailored to their specific requirements. This customisation capability is available in preview and promises to be a game-changer for developer productivity.
5. Generative Business Intelligence (BI) in Amazon QuickSight
Amazon QuickSight, a cloud-based unified BI service, is now equipped with Generative BI authoring capabilities. This functionality empowers business analysts to create customisable visuals using natural-language commands. Analysts can describe their desired outcomes, and QuickSight generates compelling visuals, reducing the time spent manually creating and adjusting charts and calculations.
For more information and to explore generative AI on AWS, visit aws.amazon.com/generative-ai/.
