Amazon Inferentia Reviews and details

Social recommendations and mentions

• Amazon spends $2.7B on startup Anthropic in largest venture investment

  > Here it says they're going to use Amazon's chips for training and inference, but...Amazon doesn't have its own chips yet??? Amazon has had its own chips for years. https://aws.amazon.com/machine-learning/inferentia/ https://aws.amazon.com/machine-learning/trainium/. - Source: Hacker News / 7 months ago

• Amazon spends $2.7B on startup Anthropic in largest venture investment

  No idea if it's any good or not, but Amazon has their own "Inferentia" chips. https://aws.amazon.com/machine-learning/inferentia/. - Source: Hacker News / 7 months ago

• Nvidia releases new AI chip with 480GB CPU RAM, 96GB GPU RAM

  You can use them today on AWS. [0] https://aws.amazon.com/machine-learning/inferentia/. - Source: Hacker News / over 1 year ago

• Sundar Pichai: An important next step on our AI journey

  Amazon has their own TPU equivalents for training and inference: https://aws.amazon.com/machine-learning/trainium/ https://aws.amazon.com/machine-learning/inferentia/ But, I really don't think this would be a limiting factor regardless. It's not as if an Amazon or Microsoft sized company is incapable of developing custom silicon to meet an objective, once an objective is identified. - Source: Hacker News / over 1 year ago

• Amazons home made server chips vs AMDs

  You are mistaken. See https://en.m.wikipedia.org/wiki/Annapurna_Labs and some of their work (specialized chips similar to Google’s TPU): https://aws.amazon.com/machine-learning/inferentia/. Source: almost 2 years ago

• Starting places to learn about GPU acceleration and hardware optimization in Deep Learning context?

  I work as an ML engineer, but have never worked on any deep learning tasks other than experimental, and I do not deal with particularly large datasets for the most part. I am familiar with the basics of the larger DL frameworks and have done some projects here and there, but I have very little understanding of the hardware side of things. I was at an AWS conference and attended a session about their then new ml... Source: over 3 years ago