Amazon.com Inc.
05/12/2026 | News release | Distributed by Public on 05/12/2026 10:48
10 AI chip terms you should know
5.
Inference
The process where a trained AI model applies what it has learned to generate outputs, like answering a question, translating a sentence, or creating an image. Training is how an AI learns, but inference is how it puts that training into practice. Every time you interact with a chatbot or receive a personalized recommendation, inference is happening behind the scenes. The requirements for inference are very different from training: speed and cost-per-query matter enormously, which is why chips optimized specifically for inference are increasingly important.
AWS powers the work of "inference" with custom chips, smart routing systems, and purpose-built infrastructure-making AI faster and more affordable.
Amazon.com Inc. published this content on May 12, 2026, and is solely responsible for the information contained herein. Distributed via Public Technologies (PUBT), unedited and unaltered, on May 12, 2026 at 16:48 UTC. If you believe the information included in the content is inaccurate or outdated and requires editing or removal, please contact us at [email protected]