AI & LLM Index
Glossary · Inference

What is VRAM?

VRAM is an inference-time concept in large language model systems. It is part of the vocabulary teams use when they compare models, design prompts or review an AI system.

Definition

VRAM is an inference-time concept in large language model 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 inference discussions, where precision matters: two teams can use the same word to mean slightly different things. In this glossary, vram is treated as a practical concept rather than a formal specification.

Why it matters

Inference terms describe how a trained model turns a prompt into output: how tokens are sampled, cached and streamed. They drive latency, cost and the consistency of responses in production.

Example

Suppose you are comparing Mistral Small 3 7B with another model for a production task. VRAM is one of the factors behind the numbers you see: Mistral Small 3 7B lists a 4,096-token context window, costs $0.89 per million input tokens and scores 44.1 on LiveCodeBench. Understanding the term helps you decide which of those differences actually matter for your workload.

How to evaluate it

  • Check whether vram 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

InferenceInference TokenInference PromptInference CompletionInference TemperatureInference Top-pInference

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 VRAM mean?

VRAM is an inference-time concept in large language model systems. It is part of the vocabulary teams use when they compare models, design prompts or review an AI system.

Why does VRAM matter?

Inference terms describe how a trained model turns a prompt into output: how tokens are sampled, cached and streamed. They drive latency, cost and the consistency of responses in production.

How is VRAM used in practice?

Suppose you are comparing Mistral Small 3 7B with another model for a production task. VRAM is one of the factors behind the numbers you see: Mistral Small 3 7B lists a 4,096-token context window, costs $0.89 per million input tokens and scores 44.1 on LiveCodeBench. Understanding the term helps you decide which of those differences actually matter for your workload.