AI & LLM Index
Glossary · Hardware and infrastructure

What is NPU?

NPU is a hardware or infrastructure concept in AI systems. It is part of the vocabulary teams use when they compare models, design prompts or review an AI system.

Definition

NPU is a hardware or infrastructure concept in AI systems. It is part of the vocabulary teams use when they compare models, design prompts or review an AI system. The term is used in hardware and infrastructure discussions, where precision matters: two teams can use the same word to mean slightly different things. In this glossary, npu is treated as a practical concept rather than a formal specification.

Why it matters

Infrastructure terms describe the memory, compute and serving choices behind a model. They explain why the same model can have very different cost and latency profiles across providers.

Example

Suppose you are comparing o4 Max with another model for a production task. NPU is one of the factors behind the numbers you see: o4 Max lists a 2,000,000-token context window, costs $19.51 per million input tokens and scores 94.6 on HumanEval. Understanding the term helps you decide which of those differences actually matter for your workload.

How to evaluate it

  • Check whether npu is defined the same way in the documentation you are reading.
  • Ask which measurement or trade-off the term is describing before comparing numbers.
  • Separate vendor claims from independently observed behaviour on your own data.
  • Write down the definition your team uses so reviews and evaluations stay consistent.

Related terms

GPUHardware and infrastructure TPUHardware and infrastructure LPUHardware and infrastructure HBMHardware and infrastructure VRAMHardware and infrastructure Memory bandwidthHardware and infrastructure

Related guides

How to choose an LLM for codingBuying How to choose an LLM for code reviewBuying How to choose an LLM for debuggingBuying How to choose an LLM for data analysisBuying

FAQ

What does NPU mean?

NPU is a hardware or infrastructure concept in AI systems. It is part of the vocabulary teams use when they compare models, design prompts or review an AI system.

Why does NPU matter?

Infrastructure terms describe the memory, compute and serving choices behind a model. They explain why the same model can have very different cost and latency profiles across providers.

How is NPU used in practice?

Suppose you are comparing o4 Max with another model for a production task. NPU is one of the factors behind the numbers you see: o4 Max lists a 2,000,000-token context window, costs $19.51 per million input tokens and scores 94.6 on HumanEval. Understanding the term helps you decide which of those differences actually matter for your workload.