Amazon has announced the availability of Meta's Llama 3.3 70B on Amazon Web Services (AWS), enabling customers to build and scale generative artificial intelligence (AI) applications. This advanced large language model expands options for deploying AI solutions on AWS.
According to a news release by Amazon, Meta’s Llama 3.3 70B, a text-only large language model, is now accessible on AWS through Amazon Bedrock, SageMaker, and EC2 with Trainium and Inferentia chips. The model offers improvements in reasoning, mathematics, general knowledge, and instruction-following compared to earlier versions and performs similarly to larger models like Llama 3.1 405B while requiring fewer computational resources. It is designed for tasks such as multilingual dialogue, text summarization, and analysis, supporting scalable generative AI applications.
Amazon Web Services (AWS) and Telecom Advisory Services have released new studies measuring the economic impact of cloud computing, AI adoption, and increased energy productivity. According to a news release by Amazon, the studies show that in 2023 cloud adoption contributed over $1 trillion to global GDP, with cloud-enabled AI generating $98 billion. They project that by 2030 cloud adoption will surpass $12 trillion in GDP, with cloud-enabled AI accounting for an additional $1.5 trillion.
AWS is described as a comprehensive cloud platform offering over 200 services, including infrastructure technologies and emerging solutions such as AI and machine learning. According to Amazon, it serves a wide range of customers including startups, enterprises, and government agencies with a broad partner network focused on providing cost-effective scalable tools across various industries.
Amazon provides various services including online and in-person shopping, delivery of fresh produce, and media distribution. The company operates devices and services such as Alexa and Fire TV while offering cloud solutions through AWS. Its operations include logistics and customer service; it also delivers entertainment through platforms like Amazon Originals, Prime Video, and Audible according to the What We Do webpage provided by Amazon.